"

7 Chapter 7: Social Implications of AI

A group of humanoid robots with sleek, futuristic designs work together on an assembly line in an advanced industrial facility. The robots, adorned with metallic exteriors and illuminated interfaces, meticulously assemble intricate components. Each has a distinct color scheme, with white, yellow, and silver variations, symbolizing specialization or different roles. Their mechanical hands handle small parts with precision, reinforcing the theme of AI-driven automation in manufacturing. Overhead lights cast a futuristic glow, emphasizing the high-tech setting.
GenAI Art 7.1 – AI Collaborations

By now, you’ve likely interacted with AI in various forms, knowingly or unknowingly. Whether it’s being greeted by a voice in a call center, having Alexa play your favorite song, or using generative AI to brainstorm ideas for your thesis, AI applications are woven into the fabric of our daily lives. As we’ve explored in previous chapters, AI’s potential to enhance our lives is vast. It provides self-driven cars for people with disabilities, discovers the next best protein sequence, or detects cancerous cells in human skin. AI has already improved our lives in ways we once thought impossible.

However, these advancements come with significant social implications and risks such as concentration of power, job displacements, ethical and regulatory dilemmas, dependency on AI, security issues and bias risks. Issues like discrimination and bias, which we’ve discussed in previous chapters, remain pressing concerns. Our earlier discussions have shown that AI still struggles with generating sexist, homophobic, racist, or xenophobic content.[1] [2] Algorithms developed for AI systems are criticized not only for the biased content embedded in their training data but also for exploiting this data without proper consent. These practices have vast social impacts. To address these issues, this chapter focuses on the intersection of AI and privacy, the impact of AI on employment, and the importance of international collaboration for safe AI implementation.

In preparing this chapter, I sought a metaphor to illustrate the intersection of AI and privacy, breaking down this complex topic into relatable concepts. ChatGPT-4o did not disappoint, offering an easy-to-understand metaphor that we can all relate to.

OpenAI ChatGPT Green Logo Certainly! Let’s extend the house metaphor to cover how AI intersects with different aspects of privacy:

Imagine your personal data is like a house.

1. Data Collection (Who Enters Your House)

  • Visitors (Authorized Data Collection): When you invite guests into your home, you’re aware of their presence and have control over their access. This represents AI systems that collect data with your consent, like a health app you use to track your fitness.
  • Intruders (Unauthorized Data Collection): If someone breaks into your home without your permission, it’s a breach of your privacy. This represents AI systems that collect data without your knowledge, like hidden tracking scripts on websites.

2. Data Usage (How Your House is Used)

  • Friends Over for a Party (Intended Use): When you host a party, your home is used as intended—people gather, socialize, and then leave. This is akin to your data being used for the purposes you agreed to, such as improving service recommendations.
  • House Party Gone Wild (Misuse): If your party turns into an uncontrolled event, with people you didn’t invite causing damage, it reflects the misuse of data—like a company selling your data to third parties without your consent.

3. Data Sharing (Who You Share Keys With)

  • Giving a Spare Key to a Neighbor (Controlled Sharing): You might trust a neighbor with a spare key to your house for emergencies. This represents sharing your data with trusted third parties, like sharing health information with a specific app for personalized advice.
  • Lost Keys (Uncontrolled Sharing): Losing your house key means anyone who finds it can enter your home. This is like data breaches where your information is exposed to unknown parties.

4. Data Storage (Where You Store Your Belongings)

  • Safe Storage (Secure Data Storage): Storing valuables in a safe ensures they’re protected. Similarly, encrypting your data and storing it in secure locations protects it from unauthorized access.
  • Cluttered Storage (Insecure Data Storage): Leaving valuable items scattered around the house makes them vulnerable to theft. In the digital realm, this is akin to storing sensitive data without proper security measures.

5. Data Retention (How Long You Keep Things)

  • Spring Cleaning (Regular Data Deletion): Regularly going through your belongings and discarding what you no longer need is like periodically deleting outdated or unnecessary data, reducing the risk of exposure.
  • Hoarding (Excessive Data Retention): Keeping every item you’ve ever owned creates clutter and potential hazards. Similarly, retaining data indefinitely can increase the risk of data breaches and misuse.

6. Transparency (Knowing What’s Happening in Your House)

  • Clear Windows (Transparency): Having windows lets you see outside and others see inside, representing transparency in how your data is being collected and used.
  • Tinted Windows (Lack of Transparency): Tinted windows obscure visibility, reflecting a lack of clarity and transparency about data practices.

Do you not think that this metaphor clarifies some of the most important privacy concerns? In this section, we explore key privacy issues and the broader societal ramifications of AI that scholars have been actively debating.

AI and Privacy Concerns

Learning Objectives

  • Explain key privacy concerns associated with AI technologies.
  • Evaluate frameworks that aim to protect individual privacy in AI.
  • Describe methods for enhancing data privacy.
  • Analyze potential ethical implications of AI on personal privacy and societal norms.

Although we previously discussed issues related to privacy in the chapter on AI ethics, there is still more ground to explore here as these technologies become integrated into our lives. Soon, many aspects of our daily lives will likely be influenced or even orchestrated by AI and robots. Unlike robots, which have a physical presence, AI is often thought of as a disembodied entity existing within the computing clouds. However, AI is deeply embedded in a different cloud—a complex mix of politics, power, and technical jargon. It is within this intricate web of political agendas and specialized terminology that privacy concerns emerge.

To address these issues effectively, it’s crucial to review some of the terminology surrounding AI to explore how they relate to privacy concerns.

A Venn diagram illustrating the intersection of three domains: Artificial Intelligence (AI), Data Protection, and Privacy Concerns. Each domain contains specific elements, with overlapping areas highlighting shared themes. Artificial Intelligence (AI): Contains "AI Algorithms," "Machine Learning," and "Predictive Analytics." Data Protection: Contains "Data Security," "Anonymization," and "Encryption." Privacy Concerns: Contains "Data Collection," "User Consent," and "Surveillance." The overlapping regions include: Between AI and Data Protection: "AI Data Security" and "AI Algorithm Privacy." Between Data Protection and Privacy Concerns: "Secure Data Practices" and "User Consent with AI." Between AI and Privacy Concerns: "AI in Data Collection" and "AI-driven Surveillance." At the center of all three domains: "Ethical AI Practices" and "Comprehensive Data Protection." This diagram emphasizes the interconnected challenges and opportunities in ensuring ethical and secure AI applications while addressing privacy concerns.
Figure 7.1 – Intersection of AI and Privacy

Data Collection and Privacy

Data privacy rules have been in place for some time. Over the years, numerous data protection initiatives have emerged. The European Union’s GDPR or information privacy law, based on the 1980 OECD Guidelines on the Protection of Privacy and Transborder Flows of Personal Data are guidelines enshrined in privacy law around the world, including the Privacy and Data Protection Act 2014 (PDP Act). The OECD principles are closely tied with the European Union legislation and cultural expectations. According to the OECD guidelines, the following represent these guidelines.[3]

Table 7.1 OECD Guidelines on the Protection of Privacy and Transborder Flows of Personal Data 
OECD Privacy Principle Description
Collection Limitation Data should be collected only with the individual’s knowledge and consent.
Data Quality Only relevant and accurate information should be collected for a specific purpose.
Individual Participation Individuals should be informed if their data is collected and have access to it if it exists.
Purpose Specification The intended use of the data must be known at the time of collection.
Use Limitation Data must only be used for the purposes specified at the time of collection.
Security Safeguards Reasonable measures should be taken to protect data from unauthorized use, destruction, modification, or disclosure.
Openness Individuals should have access to information about data collection practices and contact details of the entity collecting the information.
Accountability Data collectors should be held accountable for non-compliance with these principles, with a dedicated person responsible for adherence.

Such data privacy guidelines have been essential and worked well for years for such data/privacy intersection, but they also recognize the complexity of the intersection of data and privacy. But when it comes to AI and its intersection with privacy, additional challenges are put forth to this intersection as personal data is collected as part of AI training or queries inputted to AI. A few challenges arise with ML training. As you recall, bias is a major challenge that could be embedded within training. As we have seen in previous chapters, bias is connected with issues of fairness and transparency.

The personal data used to train machine-learning models may introduce bias into the model. Bias is tightly associated with the notions of transparency and fairness in the GDPR and is governed by employment laws like Title VII of the Civil Rights Act of 1964 and US Equal Employment Opportunity Commission Guidance on AI and Title VII. Using personal data in a manner that results in biased predictions or outputs may violate privacy and employment laws. Also, the US Federal Trade Commission (FTC) has cautioned companies that AI that is unfair or deceptive will be subject to its Article 5 authority under the FTC Act.[4]

Beyond issues of bias in collected data for training, another issue in data collection is the collection of personal data.

Sidebar – Put into Practice

Has this ever happened to you? It’s becoming increasingly common for many of us, in various ways. Imagine you’re chatting with a friend about your favorite song, “Fashion” by Lady Gaga. Your phone picks up on the word “fashion,” collects that snippet of conversation, and runs with it. Suddenly, instead of hearing more about the song, your phone starts bombarding you with news about the latest fashion trends and brands. Next thing you know, ads for clothing lines and fashion accessories are flooding your social media feeds and email inbox, even though you have no interest in them.

Before you realize it, you’re drowning in fashion ads on Facebook or Instagram—not the music you love, but products you never asked for and have no connection to. It’s as if AI has invaded your personal space, following you from the digital realm to your everyday life. It eavesdrops on your conversations, then hands over that information to marketers who bombard you with ads, shaping your online experience in ways you never consented to. How is this fair to you?

Navigating AI, Privacy, and Data Protection

The challenge at the intersection of AI and information privacy lies in the traditional reliance on individual consent for data collection thus far. But what we mean by personal consent is being redefined currently as AI models get trained on vast amounts of internet data and user interactions with AI. Social science research indicates that people often interact with AI as though it were human, complicating the notion of what constitutes consent. Traditionally, personal information refers to identifiable data about an individual. However, with technological advances, this definition now encompasses more complex and sensitive data. For example, suppose AI gathers data on a person’s mental health to assist in emergencies, including monitoring vitals to provide support in distressing situations. In that case, this information, if identifiable, becomes subject to privacy concerns based on who has access to it.[5] [6] Although the use of AI here may be new, the issues of privacy and personal health information are not new.

The intersection of AI and privacy becomes increasingly complex because of the nature of AI’s training and predictive capabilities. Since AI is trained on various data sources and can forecast outcomes, it may infer information that wasn’t explicitly present in the original input. As the Office of the Victorian Information Commissioner states, AI’s capacity to recognize patterns that the human eye cannot perceive, learn, and forecast people and groups accounts for a large portion of its value. In this way, AI can produce data that would otherwise be hard to get or nonexistent.[7] This capability means that AI might infer sensitive details about a person that they would prefer to keep private. For instance, a faculty member with a mental illness might be comfortable with their medical network being aware of their condition but not want their university to know, fearing stigma and potential job repercussions. This complexity underscores the challenges at the intersection of AI and privacy.

Consider this type of complexity in the controversy surrounding an Algerian female boxer, Imane Khelif, in the 2024 Olympics, who was accused of being biologically male. This sparked a heated debate, with figures like Paul Rogan and JK Rowling expressing strong opinions against her participation. JK Rowling in X, previous Twitter slams Paris Olympics calling it a ‘Disgrace’ after what she claims a biological male, Algerian boxer, beats a Carini, the Italian, looking female, boxer. “A young female boxer has just had everything she’s worked and trained for snatched away because you allowed a male to get in the ring with her. You’re a disgrace, your ‘safeguarding’ is a joke and #Paris24 will be forever tarnished by the brutal injustice done to Carini.”[8] Others argued that she was a woman with naturally high testosterone levels and had been vetted accordingly. The International Olympic Committee (IOC) emphasized that these references are being used to fuel hate speech. “‘We have two boxers who were born as women, raised as women, hold passports as women, and have competed for many years as women,’ said Bach. ‘Some want to define what it means to be a woman.'”[9] AI exacerbated this debate when AI-generated images allegedly proving the boxer was a male only exacerbated the controversy.[10]

Such controversy has sparked discussions on AI algorithms, misinformation, and disinformation, which may lead to rushed, unjust, unwarranted, discriminatory, and hateful conclusions. This situation illustrates how AI can perpetuate and amplify bias and discrimination via social media, potentially infringing on personal privacy and safety online.

In many ways, digital individual privacy is often framed as a simple “yes/no” consent question before using digital technology. Ideally, users can mitigate privacy concerns by rejecting the risks associated with digital platforms or voicing their concerns. However, research highlights a “privacy paradox,” in which individuals knowingly disclose personal information despite being aware of potential privacy risks. Scholars have suggested that individuals often feel they have no choice but to share their information, and some may even accept the use of their data. This paradox implies that, even when informed, individuals often feel pressured to agree to terms that they find unfair, thereby allowing their data to be used. Consequently, the binary yes/no consent option at the start of a transaction becomes somewhat meaningless.[11] [12] [13]

As AI blurs the lines between personal and non-personal information, the challenge of maintaining privacy becomes increasingly complex. As we have discussed in previous chapters, our future understanding of AI and privacy may shift toward ensuring ethical and responsible handling of data rather than just focusing on its collection. This will involve enhancing transparency and accountability throughout the AI lifecycle—from development and data collection to processing and decision-making. The ‘right to explanation’ is crucial here, allowing individuals to challenge decisions made by algorithms, especially those impacting them without human oversight, so trust between humans and machines can continue occurring.[14] Achieving this requires an interdisciplinary approach, incorporating insights from developers, regulators, governance bodies, and users. Only by reimagining traditional concepts and frameworks can we effectively build, use, and regulate AI, while addressing privacy concerns that go beyond identifiability and bias to include surveillance and broader ethical implications.

AI and Surveillance

In the era of AI technology, privacy isn’t the only concern. As the previous section illustrated, surveillance issues also loom large. AI can collect, analyze, and interpret vast amounts of data, raising stakes not only for privacy but also for the potential surveillance of people’s lives.

According to BBC, for instance, in Xinjiang, China, AI-powered surveillance is extensively employed in China, used to monitor the Uyghur population. The norm is constant security checkpoints and identification stations requiring individuals to submit digital scans, DNA samples, and facial scans when entering places and download a government phone app that collects information, including text messages and contact lists.[15] Such facial recognition and other collected data becomes data fed into local authorities’ AI-driven Integrated Joint Operations Platform (IJOP). The AI system, according to the report, is trained in detecting and analysing minute changes in facial expressions and skin pores. The system flags people it deems problematic based on collecting biometric data that helps in creating individual profiles and assess trustworthiness based on factors like familial relations and social connections. This intense surveillance system is supported by significant security spending and the deployment of advanced technologies, like surveillance drones, creating a pervasive environment of AI-driven monitoring in the region.[16] [17]

The topic of surveillance and AI’s participation in it is complex and complicated. While the majority of people would not enjoy having AI follow them around, some uses of AI surveillance may be justifiable. For instance, reexamining the home and privacy metaphor that ChatGPT offered above for installing surveillance-capable cameras. Monitoring both indoor and outdoor activity can improve security. However, this also poses risks, as footage can be stored in the cloud and potentially used to train AI for future predictions, analysis, or decision-making. Surveillance is closely tied to a complex interplay of privacy concerns, politics, and power.

In sum, the rapid developments in AI training and use have raised concerns about user consent, the ethical use of personal data, and privacy rights in general. Let’s explore how AI is trained, what some of the lesser-known risks are, and what steps can be taken to ensure the benefits outweigh them.

The Mix of Politics and Power

A modern building structure resembling Michel Foucault's concept of the Panopticon. The central focus is a tall, multi-story watchtower with narrow windows and observation decks, surrounded by multiple tiers of cells or rooms arranged in a circular or semi-circular layout. The design emphasizes surveillance, with the watchtower enabling a view of every surrounding room. The concrete walls and barred windows reinforce the institutional and controlled atmosphere, reflecting themes of observation, discipline, and power dynamics. The bright daylight highlights the starkness of the architecture.
GenAI Art 7.2 – Michelle Foucault’s Panopticon

Technological surveillance isn’t a novel concept. The metaphor of the panopticon or the panoptic tower (pan=all; optic=seeing), first proposed by 19th-century philosopher Jeremy Bentham and extensively studied by Michael Foucault, offers a useful lens. In Bentham’s model, a central watchtower allows a single guard to observe all prisoners without them knowing whether they’re being watched. This “perfect prison” design ensures that inmates must assume they’re under constant observation and cannot know when they are being surveilled. But, they know that at any moment a guard could be watching.[18] Call center labor is arguably a perfect illustration of this, as supervisors constantly listen to their staff members. Still, they are never aware of whose employee is being listened to at a particular time.

While Bentham viewed this to maintain order, Michel Foucault expanded on the idea, suggesting that modern societies have evolved into cancel culture through technological surveillance. Our movements and behaviors are constantly monitored, often from undetectable vantage points, in the name of social organization. Federal agencies can track us through the internet, telephones, social security numbers, credit cards, ATMs, and the ever-growing number of surveillance cameras in urban spaces.

In his famous book, Discipline and Punish, Foucault described the Panopticon as being polyvalent in its applications. It can reform prisoners. It also can serve to treat patients, educate students, confine the insane, and supervise workers. It is a hierarchical organization of the disposition of centres and channels of power that can hold bodies in space and the distribution of individuals in relation to one another. Such power can be implemented in hospitals, workshops, schools, and prisons. The panoptic schema can be applied whenever the people in control need to deal with individuals who need to be persuaded to perform a task or exhibit a specific behavior.[19]

Foucault argued that we already live within a panopticon-like system embedded in its mechanisms. Now, imagine how AI amplifies this, extending surveillance to previously unimaginable levels.

The Digital Eye

A futuristic surveillance device featuring four large, metallic security cameras arranged symmetrically around a central core. The structure is adorned with glowing blue circuit-like patterns, giving it a high-tech, AI-powered appearance. The cameras have an imposing presence, symbolizing constant observation and advanced monitoring capabilities. The device sits on a sleek, illuminated base, emphasizing its integration with digital and AI-driven surveillance systems.
GenAI Art 7.3 – AI Surveillance generated by AI meta

AI systems trained on vast datasets sourced from the internet carry inherent biases and discriminatory patterns, which can influence the predictions, analyses, and decision-making processes of AI systems. The proprietary nature of many AI algorithms makes them difficult to scrutinize. In an insightful Forbes article, Matthias Pfau highlights the lack of transparency in how these algorithms function and make decisions. This opacity can lead to biases that disproportionately affect certain groups, particularly minorities, under the guise of surveillance.[20]

Giddens distinguishes between two axes of surveillance. The first axis considers surveillance as the collection of coded data, leading to the pacification of nations. The second axis involves the direct monitoring of subordinates within a capitalist framework. This formulation of surveillance aligns with Foucault’s concept of the Panopticon, where mass surveillance has become a pervasive reality, giving rise to the notion of the “digital eye.” Though not a new concept, the expanding surveillance infrastructure—reminiscent of both Foucault’s Panopticon and Giddens’ two axes—creates a society in which individuals are constantly observed, blurring the lines between public safety and privacy invasion. Ultimately, with technological advancements, concerns about surveillance and data collection have grown, ultimately giving rise to the concept of the “digital eye.” Today, AI systems serve as the new “digital eyes,” continuously monitoring and analyzing nearly every aspect of human life.[21] [22]

Around the world, AI-powered systems are enabling governments to monitor the public more efficiently. For instance, in 2018, Singapore used AI systems and facial recognition cameras in lampposts for nationwide monitoring to surveil citizens for littering acts. Similarly, Malaysia partnered with China’s Yitu Technology to equip police with an AI-powered facial recognition system linked to a central database for real-time identification of citizens via body camera footage. Both Chinese and American surveillance technologies have played pivotal roles in equipping countries with tools to monitor their populations.[23]

Globally, AI-powered surveillance has become a powerful tool for regimes to bolster their authority, as evidenced by the widespread adoption of these technologies. According to the 2019 AI Global Surveillance Index, 56 out of 176 countries use AI to monitor and control their populations, often to preempt civil unrest.[24] This trend raises significant ethical concerns, particularly regarding the potential suppression of fundamental democratic practices, such as protests. Research also indicates that the intersection of AI surveillance and power suggests that the proliferation of AI technology may reinforce autocratic regimes. For example, the use of facial recognition AI has been linked to both governmental and commercial innovations, often to the detriment of dissenting voices.[25] [26]

A highly advanced humanoid robot with a metallic silver body and glowing blue eyes stands in a cyberpunk-style cityscape at night. The robot's design features intricate mechanical components, exposed wiring, and illuminated circuitry across its chest and arms. Its arms are crossed in a posture of authority, reinforcing its role as a surveillance entity. The futuristic urban background is filled with neon lights and digital billboards, evoking a high-tech, AI-driven world focused on monitoring and control.
GenAI Art 7.4 – Robot Surveillance

The surveillance issues surrounding AI highlight the need for greater awareness of AI responsibility and ethics. They also offer new ways to examine the relationship between AI, politics, and power, with a particular focus on marginalized perspectives. David J. Gunkel’s recent Handbook on the Ethics of Artificial Intelligence explores the power and politics embedded in AI, from training data to human interaction.[27] The book highlights the politics involved in the appropriation of labor, such as using artistic images or text in AI training without crediting the original creators. It also addresses the environmental impact of AI, noting that while generative AI systems are disembodied and exist “in the cloud,” they still have a physical presence that contributes to the carbon footprint. For example, the handbook points out that OpenAI’s ChatGPT consumes approximately 500 ml of water for every 5 to 50 prompts it processes.[28]

Additionally, new voices are emerging that challenge the predominantly Western understanding of AI ethics. These voices draw attention to marginalized perspectives, emphasizing the continuation of colonialism through Western norms masked as inclusivity, as noted by Min Sun Kim (2024), an author from the same handbook. Others have critiqued the whiteness of AI,[29] while others discuss the AI empire’s deep entanglement with heteronormativity.[30]

For these thinkers, achieving transparency, accountability, and explainability involves examining both the center and the periphery of these issues, from top to bottom. Addressing privacy concerns cannot be disentangled from the other three pillars of AI ethics (transparency, accountability, and explainability) throughout the AI lifecycle, particularly when surveillance methods risk infringing on individual privacy within the societal panopticon.

Protecting Privacy in an AI-Driven World

Some commentators have likened the current surge in AI development to a “Big Bang” of data or even an “AI Oppenheimer moment.” These comparisons highlight the transformative impact of AI, marked by the explosion of big data characterized by the Doug Laney’s three V’s of big data: volume, variety, and velocity.[31] According to the National Institute of Standards and Technology (NIST), volume refers to the vast amounts of data being generated, variety indicates the diverse forms of data, and velocity describes the rapid speed at which data can be processed and shared. With the rise of AI, these three Vs have been significantly amplified, leading to heightened privacy concerns resulting from AI surveillance.[32]

In the previous section, we discussed how AI algorithms can raise privacy concerns, particularly when deep learning models extract personal information from various data inputs. As AI evolves, it enhances the capability to analyze personal information with unprecedented power and speed.[33] In other words, AI systems can make inferences or predictions about individuals that may extend beyond the data they explicitly provided.

AI Surveillance Infringing on Individual Privacy

A conceptual diagram highlighting the ways AI surveillance infringes on individual privacy. The central circle labeled "AI Surveillance Infringing on Individual Privacy" is surrounded by four connected circles, each representing a specific area of concern: Facial Recognition (top): Refers to AI technology that identifies individuals based on their facial features, often raising concerns about tracking and surveillance without consent. Smart Home Devices (right): Represents AI-powered devices like voice assistants and IoT devices that collect data from personal spaces, potentially compromising privacy. Predictive Policing (bottom): Refers to AI systems used by law enforcement to predict crimes or identify suspects, which can lead to profiling and privacy violations. Online Monitoring (left): Denotes the tracking of online activities by AI, including browsing history, social media interactions, and purchases, often without explicit user consent. The diagram visually emphasizes the interconnected nature of these issues and their collective impact on individual privacy.
Figure 7.2 – Surveillance Infringing on Individual Privacy

The relationship between various AI surveillance methods and their potential impact on privacy is multifaceted. Following, we discuss training and privacy, data breach, and inadequate data anonymization.

Take the training of data, for instance, which we’ve mentioned a few times. Many people were unaware that their data from the Internet was used for training. OpenAI Inc. is facing numerous lawsuits for training its generative AI using content such as articles and videos from the internet without proper consent. For example, OpenAI was sued for allegedly transcribing millions of YouTube videos without the creators’ consent to train its generative AI software products.[34] This lack of transparency is a significant infringement on individual privacy, as people’s data is being used without their knowledge or consent. As for its legality, OpenAI (and other GenAI companies) argue that the data used to train their models falls within the “fair use” doctrine under copyright law.[35] As of Spring 2025, these lawsuits are making their way through the legal process.

Obviously, security breaches are another critical concern with privacy infringement. Cyberattacks, hacks, and the unauthorized theft of personal data are commonplace. For instance, you may have received a letter from one of your service providers informing you that there has been a breach of your personal data within their organization, resulting in the theft of your demographic or financial information. Typically, such letters contain advice on how to monitor your bank account for suspicious transactions, but the damage—the infringement on your personal information—has already been done. Not only can AI machines execute these hacks, but attackers can also manipulate input data in adversarial ML, causing AI models to make incorrect decisions. This can lead to privacy violations, data theft, and financial loss.

As you may remember from the diagram above, “the intersection of AI and different aspects of privacy,” insufficient data anonymization is another issue at the nexus of privacy and AI systems. For instance, generative AI tools may require access to personal data for training or generating outputs. If the anonymization techniques used are insufficient, it could be possible to identify individuals from the generated data. The research peer review process is a realistic example in the field of research. Data must be anonymized during this academic paper review procedure in order to protect the identity of the authors. Even though the individual in this example might not suffer any harm, consider the repercussions if data from political rallies or medical records were compromised.

As you can see in Figure 7.2, AI surveillance and privacy infringements can occur in various everyday encounters. The following figure focuses on facial recognition, online monitoring, and smart home devices, illustrating their vast social impact.

Facial Recognition

AI can collect and analyze sensitive information without individuals’ knowledge or consent, such as through facial recognition technology. As we have seen earlier, this technology is often promoted for its potential to enhance security and public safety. It allows for tracking individuals in the real world who may be defiant of agreed-upon rules from governments or groups. However, this raises significant privacy issues, including mass surveillance and the tracking of individuals in public or private spaces without their consent.

Consider, for example, the cameras installed in your school, local shopping mall, or airports. While these enhance security, they can also monitor your movements and activities without you even realizing it. The widespread use of facial recognition also raises concerns about constant monitoring and its vast social implications, such as profiling or falsifying criminal records. Imagine being wrongly accused of a crime simply because an AI system misidentified you based on your facial features. The potential for such errors underscores the need for careful regulation and oversight of facial recognition technologies.

Online Monitoring

Online monitoring involves tracking individuals’ internet activities, such as browsing history and social media interactions. This can lead to significant concerns about data privacy and the potential misuse of personal information. For instance, have you ever noticed that after searching for a particular item, ads for similar products appear on your social media feeds? This is an example of online monitoring in action.

Websites or online searches show users information that matches their previous online behavior, creating filter bubbles. If left unchecked, AI could exacerbate these filter bubbles, predicting what users might like to see and applying filters accordingly. This can limit exposure to diverse viewpoints, creating an echo chamber where only familiar ideas and perspectives are reinforced.

Smart Home and Smart Hospital Devices

Smart home devices, like smart speakers, car cameras, and appliances, collect data to enhance convenience in our lives. For example, you might ask Alexa to define a word, settle a historical debate with a friend, or find the best restaurant for your favorite food. These conveniences are luxuries that our ancestors could only dream of, yet they come with significant concerns about personal space intrusion and data access.

Imagine your smart speaker subtly eavesdropping on your private conversations or your smart camera recording every moment in your home. Who exactly has access to this data? Without your explicit consent or knowledge, this personal data can be repurposed to train AI systems, transforming these technologies into a sophisticated surveillance network that “knows” what you’re thinking, predicts your preferences, and even anticipates your actions.

The implications of data collection extend beyond the home into critical environments like hospitals. Consider the consent process when you visit your doctor or undergo a routine procedure, such as cataract surgery. In the past, you would receive a printed document to review and sign, detailing issues like HIPAA releases. Today, automation has streamlined this process. You’re often handed a digital screen with a minimal summary of what you’re consenting to, or perhaps given a brief verbal explanation from the hospital admin, leaving you in the dark about the entire contents of the document. Could you request to see the entire document? Probably. But in that moment, as you sit in the waiting room, nerves on edge for the procedure ahead, would you really take the time to ask for a printed copy? Chances are, you wouldn’t. You’re in their hands, trusting the system, even as it quietly shifts control away from you, all while relying on a binary consent model that feels increasingly obsolete in the face of such automation.

Sidebar – HIPAA and AI

When the Health Insurance Portability and Accountability Act (HIPAA) was enacted in 1996, the internet was in its infancy, and artificial intelligence (AI) was largely the stuff of science fiction.[36] HIPAA set out to protect patient health information (PHI) by establishing privacy and security rules for healthcare providers, insurers, and their partners—known as Covered Entities and Business Associates. Fast forward to 2025: AI is transforming healthcare, from diagnosing diseases to automating medical records, bringing both remarkable opportunities and unprecedented privacy challenges.

AI introduces unique concerns for HIPAA compliance. Machine learning models can inadvertently “memorize” sensitive patient data during training, leading to potential model leakage, where private information may be exposed unintentionally.[37] Even AI systems that generate synthetic medical data or clinical notes—intended to protect privacy—can accidentally reproduce identifiable patient details if safeguards are insufficient.[38]

Under HIPAA, any organization handling PHI must ensure the data is encrypted, access-controlled, and de-identified when possible. Today, AI developers—whether building diagnostic algorithms or predictive models—often act as Business Associates, binding them to HIPAA’s strict privacy and security standards. This designation extends to cloud service providers hosting healthcare AI applications and third-party algorithm developers working with PHI.[39]

The regulatory environment is evolving. In 2024, the U.S. Department of Health and Human Services (HHS) proposed updates to the HIPAA Security Rule, requiring AI-specific risk assessments, explainability measures for AI decision-making, and robust governance frameworks to manage AI-driven healthcare solutions.[40] Healthcare leaders like Mayo Clinic, which uses differential privacy for clinical AI, and Cleveland Clinic, leveraging federated learning, demonstrate how innovation can coexist with regulatory compliance.[41]

The stakes are high. A 2024 IBM study found that healthcare data breaches cost an average of $9.77 million per incident—higher than in any other industry.[42] HIPAA violations involving AI are already drawing millions in penalties, with the Office for Civil Rights (OCR) increasing scrutiny of AI deployments during compliance investigations.[43]

Yet, despite the risks, AI's potential to transform healthcare remains profound—from reducing diagnostic errors to personalizing treatment plans. Privacy-preserving techniques like federated learning, differential privacy, and homomorphic encryption offer promising paths forward. By adopting these innovations, healthcare organizations can harness AI's power while upholding their fundamental duty to protect patient privacy.

 

Predictive Policing

Predictive policing uses algorithms to forecast criminal activities based on data analysis. While this technology aims to improve public safety by predicting where crimes might occur, it can lead to biased profiling and surveillance. Critical scholars on AI ethics argue that these algorithms often target marginalized communities, reinforcing existing biases and inequalities.

For instance, consider a school where students have historically been disciplined more frequently than at others. Predictive policing algorithms might allocate more security resources to that school, resulting in increased surveillance and scrutiny of students. This could lead to over-policing and unjust treatment of the students, making them feel constantly watched and judged based on the school's reputation rather than on their individual behavior. This example highlights the ethical dilemmas and potential harms of relying on AI for policing and crime prevention.

These examples illustrate the various ways AI surveillance can infringe on individual privacy in our daily lives. Addressing these concerns requires a balanced approach, combining legal frameworks, ethical guidelines, technological solutions, and public awareness to ensure AI technologies are used responsibly and ethically.

AI Practices and Data Protection

As discussed in previous chapters of this book, ethical perspectives on AI are divided. Some groups believe that the decline of human abilities and intelligence is imminent, with AI set to take over soon regardless of ethical responsibilities. This dystopian view isn't without merit, as many safety experts from different AI companies have resigned because of concerns about the lack of safety in future AI developments.

Another group doesn't foresee the end of humanity but calls for ethical responsibility amidst the rapid pace of AI development. This latter group focuses on current actions that can be taken, such as ensuring transparency, explainability, and accountability in AI systems. They advocate for surveillance that maintains order without infringing on personal or societal privacy.

As AI systems interweave into the fabric of our daily lives, the implications for our privacy are vast and complex. These technologies possess the dual potential to empower and endanger. Therefore, establishing robust measures to protect our personal privacy is imperative. Moving beyond the binary approach to consent discussed earlier, the emerging literature suggests a multifaceted strategy tailored to the diverse phases and cycles of AI that renders AI ethical.

A harmonious combination of legal frameworks, ethical guidelines, technological solutions, and public awareness is essential to safeguarding privacy in an AI-driven world. While we have discussed such frameworks in previous chapters, it is helpful to review some of these frameworks developed in collaboration with AI.

Legal Frameworks

  • Regulations and Policies: Governmental organizations must offer regulations on privacy and data protection (e.g., Europe's GDPR).
  • International Cooperation: Privacy protection is a global issue. Creating uniform standards and practices through international cooperation and agreements to protect privacy across borders can remedy such problems.

Ethical Guidelines

  • Ethical AI Development: Companies need to create ethical guidelines that prioritize privacy and data security to ensure that AI systems respect individual privacy.
  • Transparency and Accountability: Clear accountability mechanisms and transparency from companies in relation to data collection, storage, and use of personal data need to be disclosed throughout the process of development and deployment.

Technological Solutions

  • Data Anonymization: Minimizing the risk of privacy breaches via techniques such as data anonymization and encryption.
  • AI Explainability: Ensuring explainability and transparency to elevate individual trust.

Public Awareness

  • Education and Awareness Campaigns: Educational campaigns are essential for raising public awareness about the importance of privacy and the potential risks associated with AI.
  • Empowering Users: Providing users with tools and knowledge to manage their privacy settings and control their data can empower them to make informed decisions about their digital lives.

Industry Best Practices

  • Corporate Responsibility: Adopting best practices for data protection and privacy, conducting regular audits, and complying with regulations are key to safe implementation.
  • Privacy by Design: Integrating privacy considerations into the design and development of AI systems from the outset can help mitigate privacy risks.

Key Takeaways

  • AI technologies raise significant privacy concerns, particularly with respect to how personal data is collected, used, and stored.
  • Frameworks like GDPR serve as essential guidelines to protect individual privacy and establish accountability in data usage.
  • Data anonymization and transparency are crucial in mitigating privacy risks, helping to maintain user trust.
  • The ethical implications of AI on privacy affect not only individuals but also broader societal norms, requiring responsible approaches to data handling.

Exercises

  • Privacy Policy Review: Analyze the privacy policy of a popular AI-driven app or website. Identify areas where the policy aligns with GDPR guidelines and areas where it could be improved.
  • Data Anonymization Project: Create a plan to anonymize a sample dataset, highlighting the techniques used to protect individual privacy.
  • AI Privacy Debate: Host a debate on whether companies should be allowed to collect user data for AI development without explicit consent.

The Impact of AI on Employment

Learning Objectives

  • Assess how AI is affecting job sectors, with a focus on job displacement and transformation.
  • Identify skills that are becoming increasingly valuable in an AI-driven workforce.
  • Evaluate strategies for adapting to AI-driven job displacement.
  • Discuss the ethical implications of AI in the workforce.
 Suppose you search online about jobs and AI. In that case, you're instantly bombarded with a sea of texts, videos, and images detailing how AI is transforming our world, particularly in terms of job displacement. These projections, often predicting millions of jobs lost globally, can be unsettling. The societal impact of AI is immense, especially when considering the risks it poses to employment. These concerns raise critical questions about our dependence on AI, the concentration of power within AI development, and the delicate balance required for its integration into the workforce.

In the following sections, we will discuss these concerns, exploring the broader impact of AI on society—from job displacement to the evolving nature of work and the adaptations for thriving in an AI-driven workforce. We will also consider the potential solutions that scholars and experts are discussing as we navigate the future of AI in the world of work.

AI and Job Displacement

A black-and-white photo from 1947 shows a knocker-up, an early human alarm clock, standing on a cobblestone street in Leeuwarden, the Netherlands. The man is wearing a flat cap, a light-colored jacket, and dark trousers. He holds a long stick, which he uses to tap on an upstairs window of a tall building to wake the occupants. The house has large windows, a decorative cornice, and a wrought-iron fence along the sidewalk.
Figure 7.3 - A Knocker-Up

Throughout history, technological progress has continually reshaped the employment landscape. Consider the profession of the "knocker-up" in the Industrial Revolution era. Before alarm clocks became household items, knocker-ups would walk through working-class neighborhoods in the early morning hours, using long sticks to tap on clients' windows, ensuring factory workers woke up for their shifts. In Figure 7.3, you can see a picture of a knocker-up in Leeuwarden, the Netherlands, from 1947. As affordable alarm clocks became widely available in the early 20th century, this occupation gradually disappeared—a simple yet clear example of technological displacement. With the advent of new technologies, there has always been a disruption to the workforce. Old jobs disappear and new jobs are created. The concept of job displacement refers to the process where machines, tools, or systems take over tasks previously performed by human workers, resulting in workforce changes that eliminate certain roles or significantly alter job requirements.

This pattern of displacement has repeated across centuries as innovations replace human labor in specific tasks. The mechanization of agriculture dramatically reduced farming employment, assembly lines transformed manufacturing, and digital technologies eliminated many clerical positions. In each wave, certain jobs disappear while new ones emerge, though often requiring different skills and appearing in different locations or industries.

AI job displacement refers to the phenomenon where AI technologies automate tasks previously performed by human workers, resulting in workforce restructuring, role elimination, or significant changes in job requirements.

The concept encompasses both direct displacement, where AI systems completely replace human workers in specific roles, and transformative displacement, where jobs evolve to require new skills as AI handles routine components of work. This transformation often creates what economists call "technological unemployment" - a situation where technological change outpaces the creation of new roles for displaced workers.[44]

A 2017 study estimated that 47% of US employment was at high risk of computerization.[45] The researchers focused on identifying occupations vulnerable to automation based on required skills and tasks. Also in 2017, a study conducted by the McKinsey Global Institute predicted that between 400 to 800 million people would need to transition to new jobs by 2023 because of automation, which would account for 15 to 30 percent of the global workforce.[46] In 2023, the World Economic Forum's Future of Jobs Report 2020 predicted that automation would displace approximately 85 million jobs globally by 2025, while simultaneously creating 97 million new positions - suggesting a significant reshuffling rather than net job loss. [47] [48] Work is changing, what those exact numbers will ultimately look like, we can't say with 100% certainty.

We sit here writing this in Spring 2025, so we have the latest information from the World Economic Forum and the Future of Jobs Report 2025.[49] The current prediction is that 41% of companies plan on reducing their workforce as AI automation expands by 2030. These are predictions, not facts. Getting concrete numbers of layoffs caused by AI is not easy. We know jobs are being lost because of AI, but other jobs are being created because of AI. According to data provided to CNN by ZipRecruiter, the number of AI-related jobs grew by 124% over the course of 2024.[50] In the AI world, an adage or proverb has been tossed around that says, "AI won't take your job, someone using AI will take your job." The most likely source of the original quotation is Richard Baldwin, but this is a guess because other variations have appeared.[51] In reality, people should upskill or learn how to integrate AI into their workflows. No matter what, work is transforming because of AI. This transformation raises important questions about economic inequality, workforce training, educational priorities, and social safety nets as societies adapt to AI-driven economic changes.

AI and Organizational Power

Beyond the concentration of power, a significant challenge when considering AI and employment is the question of who controls this powerful technology. While AI development is taking place globally, Western entities are gaining noticeable dominance in shaping AI's trajectory. This centralization of power presents two significant issues. First, the concentration of AI development, application, and ethical oversight within a few Western corporations and governments restricts the diversity of perspectives that could influence AI’s impact on society. This lack of decentralization risks favoring particular groups over others, potentially perpetuating existing inequalities.

Moreover, the centralization of ethical responsibility for AI within these same regions can propagate an ethnocentric worldview. Despite AI's potential to democratize technology, its ethical frameworks often reflect the values and biases of the societies in which they were created. As multiple authors in the recent Handbook on the Ethics of Artificial Intelligence highlight, this ethnocentric approach to ethics often disguises ideas of inclusivity as mere assimilation into Western norms, thereby maintaining a form of ideological colonization. Beyond reinforcing the whiteness of the field, this centralized power structure limits discourse, allowing only certain voices to shape the conversation around AI.[52] This section on ethics concludes by advocating for a shift in the narrative, urging us to move away from viewing AI through a narrow lens of “winners” and “losers.” Decentralizing AI ethics, both in terms of discourse and practice, can foster collaborative efforts essential for a technology that impacts the entire planet.

A visually striking representation of a brain divided into two contrasting halves. The right half appears organic, with detailed folds and textures resembling a natural human brain. The left half, in contrast, is futuristic and digital, featuring a translucent surface embedded with glowing blue circuits, neural networks, and data pathways. The image symbolizes the intersection of human cognition and artificial intelligence, highlighting AI-human brain dependency and technological augmentation.
GenAI Art 7.5 - AI-Human-Brain Dependency

AI’s impact on employment raises concerns about our growing dependency on these systems. As discussed extensively in education, reliance on AI can lead to plagiarism, a reduction in creativity, and a decline in critical thinking. These issues extend into the workplace as well. Some researchers explored the risks associated with overreliance on AI, identifying several ways in which AI could render our cognitive abilities obsolete: [53]

  • Reduction of cognitive effort: By outsourcing tasks like remembering appointments, performing calculations, or solving problems to AI, we may weaken our memory and attention skills, leading to greater dependence on external aids.
  • Inhibition of intuition: AI’s constant feedback might interfere with our intuitive processes—such as insight, incubation, or evaluation—thereby diminishing our creative originality and problem-solving efficiency.
  • Limitation of exploration: AI's guidance or constraints could restrict our divergent and convergent thinking processes, reducing both creative diversity and problem-solving flexibility.
A futuristic anatomical illustration of a woman's head and upper torso, partially revealing the inner workings of her brain, nerves, and circulatory system. The right side of her face remains intact with a smooth, metallic-like black surface, while the left side exposes detailed neural pathways, veins, and the structure of the brain. The intricate rendering highlights the fusion of human biology and artificial intelligence, symbolizing the integration of AI with human cognition and identity.
GenAI Art 7.6 - AI-Woman-Brain

These cognitive concerns are deeply intertwined with employment issues. Balancing AI-assisted decision-making and human input is crucial to preserving our cognitive abilities in the workplace. However, the conversation around AI and employment must also address the genuine threat of job displacement. While much focus has been on disembodied AI residing in cloud computing, embodied AI—such as robotics—poses significant risks to various types of jobs as well.

AI’s Impact on Society and Employment

Remember the futuristic worlds depicted in movies like Blade Runner or Terminator? These apocalyptic visions warned of a future where machines dominate, leaving humans to grapple with the consequences. While we may not be living out these extreme scenarios, the rapid advancement of AI is undeniably reshaping our society, particularly in terms of employment.

In those films, AI’s impact on society was catastrophic, with jobs being taken over in every sector, from warfare to daily tasks. Today, the reality of AI’s influence on employment is becoming increasingly apparent. The growth of AI across various industries has spurred many organizations to produce reports on which sectors, tasks, and even demographics are most vulnerable to automation.

The Reality of Job Displacement

Professional YouTube videos and reports paint a stark picture of AI’s potential to displace millions of jobs. One such video predicts that up to 100 million jobs, and possibly 4 billion worldwide, will be affected by AI. These explorations delve into the massive job reductions and the relentless march of automation. In this YouTube video, an AI podcaster highlights how AI is replicating not just physical tasks but also reasoning abilities, a trend that’s already visible in our daily lives—whether at the grocery store or the airport.[54]

Take, for example, the almost fully automated restaurant [55] in Pasadena, California. Robots cook and AI-powered ordering kiosks have replaced much of the human workforce which, according to the vide,o cut down on employment costs. In Asian countries, the march of automation has gone even further, with driverless taxis becoming a reality since 2023. China, in particular, is heavily investing in AI, aiming to lead the world by 2030.

Recently, Jason was walking through the Denver International Airport on a very long layover. Looking for a place to get coffee, he ran into his first fully automated robo-barista in the middle of the airport, Barista Bot. There were long lines at all of the other coffee venues, but none at Barista Bot, so he took a shot. Now, there have been coffee vending machines around for decades, but this wasn't the same thing. Jason ordered a mocha then watched as the robot created the drink just like any other barista would (see video). Many jobs will be automated. The smartest thing someone can do is learn how to integrate these new tools into their workflow instead of resisting them.[56]

While the race for AI supremacy between corporations and governments continues, one thing is clear: job displacement because of automation is increasing by the minute.

AI’s Potential Impact on the Workforce

Researchers examine the potential increase of AI in the job market; while job displacement in some industries seems inevitable, the potential of some industries with new career opportunities with AI seems real. A report by Goldman Sachs suggests that AI could replace the equivalent of 300 million full-time jobs, affecting a quarter of all work tasks in the U.S. and Europe. While this might lead to new jobs and a productivity boom, the immediate impact on the workforce is daunting. The report predicts that two-thirds of jobs in the U.S. and Europe are exposed of AI automation, with around a quarter of all jobs potentially being performed entirely by AI.[57] [58]

Researchers from the University of Pennsylvania and OpenAI found that educated white-collar workers earning up to $80,000 a year are among those most likely to be affected by workforce automation. According to a Forbes article, an MIT and Boston University report estimates that AI will replace two million manufacturing jobs by 2025. The McKinsey Global Institute predicts that by 2030, at least 14% of employees globally could need to change their careers because of digitization, robotics, and AI advancements.[59]

Gender Disparities in Job Automation

A humanoid robot with a sleek, futuristic design sits at a desk in a modern office, working on a digital tablet. Its metallic body is equipped with glowing blue circuits, and its illuminated eyes focus on the screen. In the background, a human employee in business attire observes from a distance, while multiple computer monitors display data charts and analytics. The scene represents the integration of artificial intelligence in the workforce, highlighting automation and AI's role in data analysis and decision-making.
GenAI Art 7.7 - Robot Working on a Tablet

As we already mentioned, The World Economic Forum predicted that automation would replace approximately 85 million jobs by 2025.[60] The impact is not uniform across industries or demographics. For instance, Freethink reports that up to 65% of retail jobs could be automated by that year, driven by technological advancements, rising costs, tight labor markets, and reduced consumer spending.[61] Price Waterhouse Coopers estimates that by the mid-2030s, up to 30% of jobs could be automatable, with men being slightly more affected in the long run because of the automation of manual tasks where their employment share is higher.[62] However, during the initial waves of automation, women could be at greater risk because of their higher representation in clerical and administrative roles.[63]

A 2023 study by the McKinsey Global Institute indicated that women are 1.5 times more likely than men to lose their jobs to AI.[64] The study predicts that millions of office support and customer service positions could be eliminated, affecting around 12 million people. Reports like that from McKinsey Global highlight that women and people of color, who are overrepresented in customer service and office jobs, are vulnerable to job displacement as AI reshapes the job landscape.

As AI continues to advance, its impact on society and employment is profound. While AI may enhance some industries, many others face significant challenges, particularly in terms of job displacement. The concentration of power in AI development and the resulting disparities in who benefits from these technologies require careful consideration and action.

The Changing Nature of Work

While changes are inevitable, fostering a future where AI systems enhance human potential rather than diminish it must be the priority. Adapting to the evolving nature of work and preparing for the AI-driven job market is crucial, which we will explore next.

Industries Most Affected by AI Automation

The retail industry is already feeling the effects of AI automation, with self-checkout stations becoming commonplace in grocery stores and big-box outlets. These stations result from cost-benefit analyses by companies looking to reduce labor costs. In advertising, the shift toward web and social media platforms, with their built-in target marketing capabilities, is another example of how automation is transforming industries.

Warehouse automation, too, is on the rise, with AI-powered systems locating packages, directing staff, and potentially performing mechanized retrieval and loading in the future. Insurance underwriting is another field where AI is making significant inroads. Automated systems are increasingly handling tasks that once required human underwriters, particularly those involving data analysis and application within set formulas.

Customer service and receptionist roles may soon be relics of a bygone era. With AI-powered systems like AimeReception, which can see, listen, understand, and talk with guests and customers, human interaction in these fields is becoming less necessary. AI-powered bookkeeping services offer efficient, secure accounting systems that are available as cloud-based services, further reducing the need for human employees. Similarly, AI is streamlining data analysis and research, with modern computers efficiently sorting, extrapolating, and analyzing data—tasks that once required human intervention.[65]

Here are some types of jobs that may be replaced or heavily impacted by AI soon, according to ChatGPT and other Gemini.

A humanoid robot with glowing blue eyes and metallic features stands at the forefront of a futuristic prison corridor. It wears a white jumpsuit with a black belt, mirroring the uniforms worn by rows of identical robots lined up behind it. The corridor is sterile and metallic, with bright overhead lighting and prison cells visible along the sides. The image evokes themes of automation, workforce replacement, and the impact of AI on labor, particularly in controlled environments such as prisons.
GenAI Art 7.8 - AI Prison Job Displacement
  1. Data Entry Clerks: AI-powered tools can already perform data entry tasks with high accuracy.
  2. Telemarketing and Telesales: AI-driven telemarketing systems can automatically dial numbers and play pre-recorded messages.
  3. Bookkeeping, Accounting, and Auditing Clerks: AI-based accounting software can perform tasks such as data entry, invoicing, and reconciliations.
  4. Bank Tellers and Cashiers: Online banking, mobile banking, and AI-powered chatbots are reducing the need for human bank tellers.
  5. Manufacturing Line Workers: Industrial robots and ML algorithms can optimize manufacturing processes.
  6. Taxi Drivers and Chauffeurs: Self-driving cars and ride-hailing services may replace human drivers.
  7. Retail Salespersons: E-commerce, AI-powered retail platforms, and automated retail stores may reduce the need for human sales associates.
  8. Fast Food Cooks and Preparers: Fast food chains are introducing automated kitchen equipment and self-service kiosks.
  9. Truck Drivers: Self-driving trucks and AI-powered logistics systems may revolutionize the transportation industry.
  10. Customer Service Representatives: Chatbots and virtual assistants are already handling customer inquiries and support tasks.
A humanoid robot with a sleek, reflective metallic body and glowing blue eyes serves as a waiter in a high-end restaurant. It wears a white apron and a name tag, carrying a tray with a teapot, cup, and condiments. The background features a softly lit dining area with shelves of glassware and bottles, creating a warm and elegant atmosphere. The image illustrates the growing role of AI in the service industry, highlighting automation in hospitality and customer service.
GenAI Art 7.9 - AI Wait Staff

Keep in mind that while AI may replace some jobs, it will also create new ones, such as AI developers, data scientists, and AI ethicists.

Industries Resilient to AI

Despite AI's widespread impact, some industries are likely to be enhanced rather than replaced by automation. According to the aforementioned reports, human skills and emotional intelligence remain irreplaceable in healthcare, education, and the arts. In healthcare, for instance, AI is being used to improve diagnostics and surgical precision, but the human touch is still crucial, particularly in patient care and mental health.

Assisting Healthcare Professionals

A smiling doctor in a white lab coat and blue scrubs holds a tablet displaying a 3D-rendered anatomical image of a human organ. She stands in a modern medical office with a large digital screen behind her, showing a detailed diagram of the human muscular system. The environment is well-lit and professional, emphasizing the integration of AI-powered medical tools in healthcare. The image highlights how AI assists medical professionals in diagnostics, education, and patient care.
GenAI Art 7.10 - AI Help in Healthcare

In 2016, British computer scientist and Turing Award winner Geoffrey Hinton famously stated that it is time to stop radiology training. Deep learning will clearly surpass radiologists in performance within five years. This bold prediction sparked a wave of discussion, financial investment, and rapid development aimed at creating the most advanced AI radiology systems.

However, despite the excitement, the anticipated dominance of AI in radiology has not materialized as expected. The fear and confusion surrounding the potential replacement of radiologists by AI have been tempered by the reality that AI-driven radiology faces significant challenges—particularly in terms of accuracy and bias detection.[66] [67] Deep learning models require vast amounts of annotated data to train and validate their performance. Yet, in the field of radiology, obtaining such data is fraught with ethical, legal, technical, and practical challenges. Medical images, being sensitive personal data, must be meticulously protected and anonymized.[68]

Adding to the complexity is the inherent heterogeneity and diversity of medical images. Factors such as modality, protocol, device, patient population, disease type, and stage can vary widely. This variability means that models trained on one dataset may not generalize well to other datasets or scenarios, potentially leading to poor performance or errors. Therefore, while AI holds promise in radiology, its application is far from straightforward, and the role of human radiologists remains crucial.

In the mental health field, these technologies still fall short. While a plethora of new AI tools for mental health are emerging, offering features like daily affirmations or instant access to healthcare professionals during moments of distress, a critical area is still missing: emotional intelligence. The human touch remains irreplaceable, particularly for individuals grappling with mental health challenges. In these contexts, AI systems are best utilized as assistants or collaborators, augmenting human care rather than replacing it.

A humanoid robot with a sleek metallic body and glowing blue eyes leans forward in a hospital room, engaging in a serious exchange with an older female doctor wearing a white lab coat and stethoscope. The doctor, with a focused and slightly concerned expression, sits beside a hospital bed with medical monitors displaying vital signs and brain scans in the background. The scene suggests a discussion between AI and human medical professionals, reflecting the role of robotics in healthcare, diagnostics, and patient care.
GenAI Art 7.11 - Robot in a Hospital

Surgeons, for instance, rely on years of experience, knowledge, and skill to make split-second decisions during operations—elements that AI cannot fully replicate. In China, for instance, the challenges of an aging population have spurred the increased use of AI in healthcare, from detecting vitals via smartphone cameras to employing robotics for companionship and medical care. However, despite these technological advancements, the role of human caregivers remains essential. For example, at the 10th China International Senior Services Expo in Beijing, AI robots designed to support the elderly are being showcased as part of the future of elder care.[69] These innovations include robots that assist individuals with low mobility, helping them stand and move more easily. Yet, even with such tools, the human element in caregiving continues to be irreplaceable.

Augmenting Education

A young girl wearing a yellow school uniform and a blue backpack engages in a conversation with her teacher in a colorful classroom. The teacher, a woman in a white blouse, smiles warmly while gesturing toward a laptop on the desk. The background features other children working at their desks, educational posters, and bulletin boards decorated with student artwork. The scene represents AI integration in education, illustrating how technology supports personalized learning and student-teacher interactions.
GenAI Art 7.12 - AI Helping in Education

In recent years, a plethora of discussions have emerged around the role of AI in higher education, particularly with the advent of tools like ChatGPT and other generative AI technologies. These advancements have sparked debates among educators and the public alike, with both advocates and critics weighing in on this new form of learning.

If you recall, GenAI, such as ChatGPT, is designed to create “realistic” text, images, or even video based on ML algorithms. This technology has been applied to a variety of fields, including personal and educational tasks, as well as complex endeavors like chemical research for medicinal purposes. From answering questions and coding to translating languages and composing academic papers—or even crafting poetry and film scripts—ChatGPT’s capabilities are undeniably impressive.

However, these capabilities have also created a wave of concern in the educational sector. Many educators are apprehensive, with some rejecting the use of AI outright, while others are cautious about fully embracing it within higher education. The fear is that AI might undermine the traditional educational process by making specific skills obsolete. Still, there is also concrete fear that AI will take over and replace educators' jobs.

But despite these concerns, AI has proven to be a powerful tool in education. Offering opportunities to personalize learning, automate administrative tasks, and provide timely feedback to students is part of these new-changing abilities. AI-powered educational tools, including intelligent tutoring systems, are becoming increasingly popular, enhancing the learning experience in ways previously unimaginable.

A humanoid robot with a sleek white design and glowing blue eyes assists in a classroom, walking between students seated at their desks. The students, wearing school uniforms, are engaged in their work, writing in notebooks. At the front of the classroom, a human teacher explains a lesson while standing next to a digital smartboard displaying complex diagrams and equations. The scene illustrates the integration of AI-powered robotic assistants in education, enhancing learning through technology.
GenAI Art 7.13 - Robotic Teacher's Aid in the Classroom

Generative AI systems, in particular, or AI-powered systems (e.g., virtual reality, augmented reality, etc.), are not here to replace educators but rather to redefine their roles. The real value of university lecturers —something AI can do quite well, especially as it improves its hallucinations in the new iterations—but in helping students contextualize, systematize, and critically evaluate that information. In this way, educators play a crucial role in guiding students to develop ethical thinking and deep understanding, ensuring that AI enhances, rather than diminishes, the educational experience and the overall role of the educator.[70]

In the realm of education, even as AI technologies advance, the need for human presence remains undeniable. Automation may streamline specific tasks, but the complexities of maintaining, updating, and improving sophisticated software and hardware systems often require human intervention. For instance, a Computer System Analyst—a profession in high demand in recent years—plays a vital role in reviewing system capabilities, controlling workflows, scheduling improvements, and increasing automation. This is a clear example of how human expertise continues to be essential, even in a world increasingly influenced by AI.

Supporting Artists

While education is a critical factor in the discussion of AI and job displacement, particularly regarding the fear of being replaced, another equally important area to consider is the intersection of AI and the arts. Education and AI may seem distinct, but they share a common thread: both require human oversight, creativity, and the ability to adapt.

A smiling artist with wavy dark hair and a neatly trimmed beard sits at a cluttered wooden desk in a creative studio, working on a vibrant painting. He wears a casual white button-down shirt and holds a paintbrush while blending colors on a canvas. Surrounding him are various art supplies, including brushes, sketchbooks, and a laptop displaying a digital artwork. The background features an inspiration board filled with sketches and references. The scene illustrates the integration of AI in art, where traditional techniques and digital tools coexist in the creative process.
GenAI Art 7.14 - AI Help in Art

While there has been much fear surrounding AI's potential to replace human creativity, the reality is more nuanced. The magic and thrill of creating art, particularly through words, remain firmly within the domain of human competition. AI may serve as a tool or collaborator in the artistic process, but it is not a replacement for the human touch.

A YouTube video about artistia and AI orchestrated by the Museum of Modern Art (MoMa) offers a valuable perspective on the intersection of AI and the arts.[71] It emphasizes collaboration between AI and artists rather than viewing AI as a replacement. The video underscores that, despite advancements in AI, the essence of creativity and critical thinking will always require human involvement. In the discussion, podcasters engage with three artists—Kate Crawford, Trevor Paglen, and Refik Anadol—who explore how AI and ML algorithms are driving new approaches to art-making
A digital artist sits in a well-lit studio, surrounded by traditional painting supplies and digital tools. She is working on a painting while a large screen in front of her displays a futuristic AI-generated portrait of a humanoid robot with vibrant, artistic patterns. Another screen on her desk shows multiple digital artworks in progress. The workspace is filled with brushes, canvases, and sketches, illustrating the fusion of traditional artistry and AI-assisted digital creation.
GenAI Art 7.15 - AI-Enhanced Creative Process

Professor, artist, and author of "Atlas of AI," Kate Crawford, believes that the world is entering what she terms the generative turn. The shift is happening in everything from publishing to film directing to illustration worked is about to shift drastically.

Trevor Paglen has been mining data sets used to train the ML systems that surveil our daily lives. He investigates the dangerous oversimplification inherent to these processes and the ethics of the intentions behind them. He claims that unlike the tradition of computer science or engineering data, artists bring thousands of years of contemplation about the meaning of an image to the celebration. Here, artists are contributing voices to the discussion that, in my opinion, is quite urgent.

Refik Anadol sees AI as a tool available to artists. His interest is in machine learning algorithms that aren’t strictly monitored by humans. For Unsupervised, he asked how a machine, if it had only MoMA’s collection data for knowledge, would parse the history of modern art on its own. And, as an autodidact, what kind of art would it create?These three prescient thinkers are joined by curators Paola Antonelli and Michelle Kuo, who give historical context to the existential questions at play in this emerging landscape and share insights into where art might bring AI next.[72]

Ultimately, while addressing issues of job displacements and the changing nature of work in an AI world is crucial, it is equally important to consider the broader societal implications of AI, particularly in the realm of preparing for this AI-driven job market. The transformative power of AI is reshaping job markets and altering the nature of work. As we transition to the next section, we will explore the required preparations for the AI-driven job market. Understanding these dynamics is essential for fostering a workforce that can adapt to the rapid advancements in AI while ensuring that the benefits of technology are equitably distributed.

Preparing for the AI-Driven Job Market

As we wrap up this section on preparing for a future where AI dominates the workforce, it's essential to focus on a few key strategies. These strategies will help individuals navigate and thrive in an AI-driven job market. Key areas to prioritize include continuous learning, ethical considerations, and leveraging uniquely human skills.

Table 7.2 - Upskilling in the AI Economy
Category Key Areas Subtopics
Continuous Learning and Skill Development
  • Understanding
  • Skills
  • Agility
  • Stay updated on AI developments
  • Learn new AI tech
  • Adapt to new AI tools
Building a Strong Professional Network
  • Networking in AI Community
  • Mentorship
  • Interdisciplinary Knowledge
  • Engage with AI experts
  • Seek mentors
  • Combine AI with other fields,
Focus on Creativity and Critical Thinking
  • Leveraging Unique Human Skills
  • Innovation and Ideation
  • Human-AI Collaboration Skills
  • Cultivate emotional intelligence and creativity
  • Generate new ideas and solutions with AI assistance
  • Develop skills to work alongside AI
Ethics and Responsibility
  • Understanding AI Ethics
  • Promoting Responsible AI
  • Fighting the Good Fight
  • Familiarize yourself with bias, fairness, and societal impact
  • Advocate for fair, transparent, and ethical AI development
  • Address AI biases, discrimination, and ethical concerns
Understanding Global AI Policies and Trends
  • Staying Informed on AI Regulation
  • Global AI Trends
  • Industry-Specific Policy
  • Keep up with global AI policies
  • Monitor AI adoption and investment trends globally
  • Develop AI solutions aligned with industry regulations

Continuous Learning and Skill Development

In an AI-dominated workforce, the ability to learn continuously and adapt quickly is no longer optional—it's essential for career survival. As AI capabilities grow at an accelerating pace, professionals must develop a structured approach to staying current with technological advancements. This means not just passively observing AI developments, but actively engaging with new tools, methodologies, and platforms as they emerge. The most resilient careers will belong to those who view learning as a lifelong commitment rather than a periodic necessity.

  • Understanding: Familiarize yourself with the latest developments in AI, whether from leading AI corporations or emerging trends in your specific field. Understanding where AI is heading in your industry is crucial. For example, if you work in mental health, stay updated on the newest AI tools that can assist in patient care, such as AI-driven therapy bots or diagnostic tools. If you're in education, keep an eye on AI technologies that enhance learning experiences, like personalized learning platforms or AI tutoring systems. This means acquiring relevant skills and being agile in adopting new AI systems and platforms as they develop.
  • Skills: Continuously invest in your education, especially concerning AI. Learn how to collaborate with AI systems. Embrace new skills, such as learning programming languages like Python or mastering prompt engineering, which is increasingly important with the rise of LLMs. Keeping your technical skills sharp by understanding how these LLMs work or diving into emerging AI technologies like deep learning, neural networks, and AI ethics will make you more competitive and effective in your role.
  • Agility: Cultivate an open mindset toward learning and adapting to new tools, platforms, and methods. The AI field is dynamic and fast-paced, and those who can adapt quickly will have a significant advantage. While the idea of constant learning might seem overwhelming, think of it as an opportunity to keep your brain sharp. Embracing agility not only helps you stay relevant externally in your career but also fosters internal growth by continuously challenging and expanding your cognitive abilities.

Sidebar - The Future of Work: Navigating AI-Driven Job Market Changes

Picture this: self-driving trucks delivering goods, AI chatbots handling customer service, and algorithms designing entire marketing campaigns. As artificial intelligence continues to advance, the job market is undergoing a profound transformation—and fast. By 2025, studies suggest AI could automate tasks equivalent to millions of jobs, particularly in repetitive or data-intensive fields like manufacturing, retail, and even some white-collar professions.[73]

Yet this isn’t just a story of job displacement; it’s one of job transformation and creation. The World Economic Forum estimates that while AI may displace 85 million jobs by 2030, it could also generate 97 million new roles—including AI trainers, data ethicists, and human-AI collaboration specialists. [74]

In many industries, AI is already augmenting human work rather than replacing it outright. In healthcare, for example, AI handles routine diagnostics, freeing up medical professionals to focus on complex cases and patient care—areas where human skills remain irreplaceable.[75] The most resilient careers will continue to rely on distinctly human capabilities: critical thinking, creativity, emotional intelligence, and ethical judgment.[76]

However, the impact of AI on employment isn’t evenly distributed. Rural workers often lack access to reskilling opportunities, older employees face steeper learning curves, and individuals without post-secondary education are at greater risk of being left behind. [77] Addressing these disparities requires targeted strategies that go beyond one-size-fits-all solutions.

History offers perspective. The Industrial Revolution eliminated many manual trades but created entirely new industries and professions in its wake. Today, preparing for AI-driven change demands a proactive approach: identifying transferable skills, earning micro-credentials in emerging fields, and adopting a mindset of lifelong learning. Online platforms like Coursera, edX, and LinkedIn Learning now offer affordable AI literacy programs that are accessible to anyone—no need to wait for formal institutional initiatives. [78]

The message is clear: the future of work belongs to those who adapt and evolve alongside new technologies. Whether we bridge the AI skills gap—or allow technological change to outpace our ability to keep up—will depend on the choices made by individuals, businesses, and policymakers today.

 

Building a Strong Professional Network

Navigating the AI revolution cannot be done in isolation. As traditional career paths fragment and new opportunities emerge at the intersection of AI and various disciplines, your professional connections become increasingly valuable. A robust network serves not just as a source of job opportunities but as a crucial information exchange that helps you expect changes, identify emerging skills and needs, and discover collaborative possibilities that might not be visible through formal channels.

  • Networking in AI Communities: Engage actively with AI communities within your field. Networking with colleagues who are knowledgeable about AI tools and developments can open up valuable opportunities. For example, if you are a professor, consider inviting AI experts to your classes for teaching-learning exchanges. Attend AI conferences relevant to your domain; if you're in the arts, participating in AI conferences alongside art-specific ones can broaden your perspective. Additionally, writing articles that explore the intersection of AI and your field, such as AI and art, can deepen your understanding and allow you to share insights with others. Networking is crucial—it can lead to new opportunities and a better understanding of the ever-evolving AI landscape.
  • Mentorship: Seek out mentors who can help you stay informed about the latest AI developments. Mentors don’t have to be traditional; they can be YouTubers, podcasters, or other content creators who focus on AI and regularly discuss new tools and trends. These mentors can provide guidance, share insights, and help you navigate your career path, ensuring you stay ahead of the curve in AI. For instance, podcasters like Matt Wolfe discuss issues such as How AI is Changing the World [79] Emad Mostaque discussion with podcaster Tom Bilyeu on how AI Will Displace These Jobs In 3 Years! Do This To Get Ahead While Others Panic,[80] are only a few interesting podcasts that can provide you with some beginning thoughts on this field and its impact on society.
  • Interdisciplinary Knowledge: Develop skills that blend AI with other fields, such as healthcare, finance, or law. Understanding how AI applies to specific industries can significantly enhance your expertise. For instance, if you're a computer science student, go beyond just learning programming languages—consider taking a course in AI ethics or AI's impact on society. Similarly, if you're studying radiology, explore courses that cover AI in radiology and general AI tools. This interdisciplinary approach will make your skill set more comprehensive and valuable in the job market.

Focus on Creativity and Critical Thinking

Although AI excels at processing data and performing defined tasks, human creativity and critical thinking remain irreplaceable components in problem-solving and innovation. As automation takes over routine aspects of work, the most valuable professionals will be those who can think abstractly, connect seemingly unrelated ideas, apply ethical judgment, and navigate ambiguity—areas where current AI systems show significant limitations. Developing these distinctly human capabilities provides both job security and the opportunity to leverage AI tools for enhanced productivity.

  • Leveraging Unique Human Skills: One area where humans continue to have an advantage over AI is in emotional intelligence. Cultivating emotional intelligence through active listening, empathy, critical thinking, and understanding how to apply the right emotions in different situations will remain valuable. For example, in healthcare, a doctor’s ability to empathize with patients and understand their emotional needs is something that AI cannot replicate. This human touch, combined with AI-driven diagnostic tools, can lead to better patient outcomes. Enhancing creativity and problem-solving skills also allows you to collaborate with AI to develop innovative solutions, rather than being replaced by it. Soft skills, such as effective communication, will become increasingly important in a workforce where routine tasks are automated.
  • Innovation and Ideation: Focus on generating new ideas and innovative solutions—areas where AI can assist but not independently create. As AI takes over specific tasks, it's vital to continuously reskill or upskill to move into roles that AI cannot fully automate. This includes developing soft skills like empathy and human-machine communication. For example, as AI becomes more integrated into customer service roles, those who can blend AI with human empathy to handle complex customer interactions will be in high demand.
  • Human-AI Collaboration Skills: Develop skills that enhance your ability to collaborate with AI systems. Understanding how to work alongside AI effectively can significantly boost productivity and innovation. Be prepared for roles where AI augments human capabilities rather than replaces them. For instance, starting a business that leverages AI to address global challenges, such as AI-driven solutions for climate change, can position you at the forefront of impactful innovations. Understanding how to use AI tools in your job, such as AI-powered data analysis platforms, can also give you a competitive advantage in your field.

Ethics and Responsibility

As AI systems become increasingly integrated into critical decision-making processes across society, ethical considerations move from theoretical discussions to practical imperatives. The algorithms that power AI applications reflect the values, biases, and priorities of their creators and the data used to train them. Professionals who understand these ethical dimensions and actively work to ensure responsible AI development will play a crucial role in shaping technologies that enhance human welfare rather than undermine it.

  • Understanding AI Ethics: A significant focus of this discussion is the ethical considerations surrounding AI. It’s crucial to familiarize yourself with issues of bias, fairness, and the societal impact of AI technologies, particularly in your field. For example, consider the ethical implications of using a robot to assist elderly individuals in a daycare setting. Is it better for the robot to handle routine tasks while a human provides emotional support through storytelling, listening, or physical comfort, such as hugs? This approach ensures that older adults experience human connection, which a robot cannot fully replicate. Another example involves AI systems that generate data about criminal activity—if the data used to train these systems is biased, it might unfairly target marginalized groups. Understanding how these biases arise and their consequences is essential. Additionally, consider issues of fairness related to privacy, safety, and surveillance, especially when AI is used in ways that could lead to excessive monitoring of society. This knowledge is increasingly important as AI systems become more integrated into decision-making processes.
  • Promoting Responsible AI: Advocate for and practice responsible AI development, ensuring that AI applications are designed to be fair, transparent, and beneficial to society. Engage in user training, testing, or other phases of the AI development cycle to help maintain these ethical standards. For instance, if you’re involved in the development of an AI tool, actively participate in its testing phase to ensure it performs fairly across diverse user groups and doesn’t reinforce harmful biases. Being part of these processes helps ensure that AI systems are developed with responsibility and care.
  • Fighting the Good Fight: Speak out when AI perpetuates issues like systemic racism, colonialism, cultural hegemony, or various forms of discrimination (e.g., sexism, homophobia). Emphasizing intersectionality, cultural empathy, and marginality in AI design, development, deployment, and application can help establish a more inclusive and diverse approach to AI ethics—one that is not solely reliant on Western moral theories. For instance, if you’re a journalist, report on instances where AI systems are biased or discriminatory, highlighting how these technologies may reinforce societal inequalities. Public awareness and accountability are crucial in driving the demand for more ethical and responsible AI development.

Understanding Global AI Policies and Trends

The development and deployment of AI technologies are increasingly shaped by complex regulatory frameworks that vary significantly across regions and industries. As AI touches more aspects of society, governments worldwide are establishing policies to guide its use and mitigate potential harms. Professionals who understand the regulatory landscape can better navigate compliance requirements, anticipate market shifts driven by policy changes, and identify opportunities created by new legal frameworks.

  • Staying Informed on AI Regulation: Keep up-to-date with global AI policies, regulations, and standards, as understanding the legal landscape is crucial for developing compliant and ethical AI solutions. For instance, familiarize yourself with the Organisation for Economic Co-operation and Development (OECD) AI Principles, which set international standards for AI transparency, human-centered values, accountability, and the explainability of AI algorithms. These principles can guide the ethical use of AI in your industry. For example, if you’re involved in healthcare, ensuring that AI tools comply with these principles can help maintain patient trust and uphold ethical standards in medical decision-making.
  • Global AI Trends: Monitor global trends in AI adoption and investment, as these will shape the future job market and opportunities in the field. Be aware that AI development isn’t limited to Western trends—other regions may adopt different approaches and policies. For example, China has been investing heavily in AI, focusing on areas like facial recognition and smart cities, which have different regulatory and ethical implications compared to Western standards. Understanding these global variations and the risks associated with AI development, such as fairness and bias, is crucial. International collaboration is essential to address these risks and ensure that AI development benefits everyone. For example, if you work in tech, keeping an eye on how different countries regulate AI can help you anticipate challenges and opportunities in global markets.
  • Industry-Specific Policy: Focus on developing AI-driven products or services that align with both global trends and the specific regulations of your industry. Different sectors face unique risks and regulatory requirements. For instance, in education, policies might emphasize data protection, security, and ethical use of AI, ensuring that AI tools support learning without compromising student privacy. Education policy might also define what constitutes appropriate AI use, provide examples of misuse, and outline consequences for violations. If you work in healthcare, you’d need to focus on patient data privacy, informed consent for AI use, and ensuring that AI systems enhance rather than replace human judgment.

Key Takeaways

  • AI significantly influences job markets, both by displacing certain roles and creating new opportunities in technology-driven sectors.
  • Skills like critical thinking, creativity, and emotional intelligence will gain importance as they are less likely to be automated.
  • Adaptation strategies, including reskilling and upskilling, are essential to mitigate the negative effects of AI on employment.
  • Ethical concerns related to AI’s impact on employment must be addressed to ensure fair distribution of AI’s benefits.

Exercises

  • Job Market Analysis: Research how AI is impacting a specific industry, such as healthcare or finance, and present findings on potential job transformations.
  • Skills Development Plan: Create a personal development plan that includes the skills most likely to be in demand in an AI-driven workforce.
  • AI Ethics Roundtable: Discuss the ethical implications of AI-driven job displacement and propose policies that could protect vulnerable workers.

International Collaboration for AI Safety

Learning Objectives

  • Explain the importance of international collaboration in establishing AI safety and ethical standards.
  • Identify key global initiatives and organizations involved in promoting AI safety.
  • Analyze the role of shared guidelines and regulations in preventing AI misuse and promoting ethical AI.
  • Discuss the challenges and benefits of creating a unified international approach to AI safety and regulation.

If you’re as puzzled by GenAI Art 7.16 as Sanae was, don’t worry—you’re not alone. When she asked Meta AI Llama 3.1 to generate a depiction of human neurons influenced by AI, the result was this eerie visual. Intrigued, she asked the AI to explain its creation, and here’s what it had to say:

A highly detailed, futuristic visualization of an artificial neural network, resembling an organic brain with intricate, interconnected structures. The image features spherical clusters covered in branching tendrils, glowing nodes, and vibrant molecular-like components in shades of blue, pink, and gold. A central structure is adorned with a golden spire, symbolizing intelligence or advanced computation. The surrounding clusters appear to communicate through luminous pathways, representing the complexity of AI-driven neural processes and data transmission.

GenAI Art 7.16 - AI-Brain

Meta Logo The picture I generated depicts a whimsical scene where human neurons, represented as colorful, glowing cells, are decorating a futuristic space. The neurons are adorned with tiny lights, garlands, and ornaments, symbolizing the idea that artificial intelligence (AI) is enhancing and beautifying human thought processes.

In this imaginative scenario, AI is not replacing human neurons but rather augmenting and enriching their capabilities. The decorations represent the potential benefits of AI, such as:

- Enhanced cognitive abilities

- Improved memory and learning

- Increased creativity

- Augmented problem-solving skills

The image suggests a harmonious collaboration between human intelligence and artificial intelligence, where AI is elevating human thought processes to new heights."

So Sanae asked in her usual interactive manner with AI, ‘But I asked you to generate a picture where AI deteriorates human neurons,’ and yet, you gave me something that enhances them. Why?

A surreal and unsettling visualization of an organic brain-like structure intertwined with artificial neural elements. The mass appears overgrown with tangled, fleshy lobes while glowing blue energy pulses from its core, resembling an AI-driven neural takeover. Delicate tendrils and digital synapses extend outward, hinting at an invasive influence. The dark background enhances the eerie contrast between organic matter and artificial intelligence, symbolizing concerns over AI's potential negative impact on human cognition and mental processes.

 

Meta Logo I understand now that you wanted an image depicting AI having a negative impact on human neurons, but I generated the opposite. I strive to generate images based on the input, but sometimes I may misinterpret or not fully understand the context.

To better meet your request, I can try generating a new image that represents AI having a detrimental effect on human neurons, such as:

- Neurons shrinking or fading away

- AI tendrils or circuits overwhelming or suffocating neurons

- Neurons becoming tangled or disconnected due to AI influence

Please let me know if you would like me to generate a new image based on this understanding.

So, again, she asked Meta AI Llama 3.1 to explain this picture.

Meta Logo The picture I generated depicts a disturbing scene where artificial intelligence (AI) is harming human neurons, represented as colorful, glowing cells. The AI is shown as a dark, mechanical entity encroaching on the neurons, causing them to:

  • Shrink and fade away
  • Become tangled and disconnected
  • Overwhelmed by AI tendrils or circuits

This image symbolizes the potential risks and negative consequences of AI on human cognition, such as:

  • Loss of creativity and original thought
  • Decreased memory and learning capacity
  • Dependence on AI, leading to diminished human intellectual capabilities

The dark, ominous tone of the image represents the concerns and fears surrounding the impact of AI on humanity's cognitive abilities and individuality.

While this is a chilling interpretation from AI about AI, highlighting the fears many of us share about AI’s potential impact on our minds, our encounters with AI are everywhere these days. For years, headlines have promised that AI will revolutionize our daily lives, and much of this book discusses these AI-driven transformations. Yet, as this image and its unsettling explanation remind us, there’s a darker side to AI’s incredible promise. The question isn’t just what AI can do but what it might do to us—and how we can regulate it.

In this section, we’ll discuss AI regulation, exploring how we can shape this technology to build a fairer, safer future. While discussions about AI’s rights and its place in our world are emerging, our focus will be on how governments and institutions can guide AI development, ensuring it serves humanity rather than controls it. We’ll touch on the current debates about AI rights, but our spotlight will be on AI governance—crafting the rules that keep AI on the right path.

By the end of this section, you’ll feel more empowered and less overwhelmed by the complexities of AI regulation. We’ll address some concerns about balancing AI’s incredible potential with the safeguards needed to keep it in check, helping you envision a future where AI and humanity thrive together.

The Need for Global AI Governance

As you know, AI is the talk of the town, and for good reason. Promising to revolutionize industries, transform our daily lives, and even outsmart humans in specific tasks is part of the continuous talks about AI. Some people see AI as the ultimate solution to long-standing problems, from diagnosing diseases to composing music, driving cars, and even beating world champions at complex games. For example, if you’ve ever wondered whether your dog’s bark is playful or aggressive, AI’s got that covered, too. The University of Michigan has developed an AI tool that not only interprets your dog’s bark but can also determine its sex, age, and breed.[81]

However, the conversation doesn’t stop at the benefits. You also hear about the challenges AI brings, such as its significant environmental impact, like consuming vast amounts of water, which can contribute to climate change. Moreover, AI can be biased, raise privacy concerns, and even be discriminatory against certain societal groups. Given these complexities, you might start to wonder: Is there a way to harness AI’s power while keeping it in check, ensuring that it remains fair, safe, and transparent? How do we regulate AI to ensure it aligns with our values, and not the other way around?

As AI continues to become a staple of our daily lives, a parallel movement has emerged to ensure that this powerful technology remains aligned with human values. While the design, development, and deployment of AI systems dominate discussions—especially in tech circles and podcasts—a growing concern has prompted the rise of Responsible AI and AI governance. These concepts have become buzzwords, driving reports, scholarship, and collaborative efforts across various sectors. But while "responsible AI" and "ethics" are often clearly defined, AI governance can seem murky, with definitions and details becoming entangled in a web of academic jargon and industry reports. For simplicity, let's define AI governance as:

A set of rules, standards, and processes that guide AI systems to maximize their benefits and mitigate their risks. In essence, AI governance acts as the backbone of any AI initiative, ensuring that these systems don’t just function, but thrive in a way that aligns with societal values.

A two-panel composite image depicting AI integration in corporate governance. The top panel shows a busy open-office environment where humanoid robots in business attire interact with human employees, working at desks with computers and paperwork. The bottom panel presents a closer view of a bearded businessman in a vest and tie sitting at a desk, taking notes while having a formal discussion with a humanoid robot. The robot, designed with a metallic frame and expressive artificial eyes, appears to be engaged in decision-making. The scene symbolizes AI's role in corporate strategy, policy-making, and governance, highlighting the collaboration between humans and artificial intelligence in administrative and executive functions.
GenAI Art 7.17 - AI Governance

Without a well-thought-out governance plan that brings together all aspects of AI—tools, data, people, and regulations—the risk of failure is inevitable. AI governance is not just a safety net; it’s the blueprint that ensures every component, from the technology itself to the people who create and use it, works together harmoniously.

A valuable resource, the AI Glossary of Credo AI, offers core aspects of AI governance that can significantly enhance our understanding and discussion.[82]

The Core Aspects of AI Governance

As artificial intelligence systems become increasingly embedded in critical infrastructure and decision-making processes across society, the need for robust governance frameworks has moved from theoretical to imperative. AI governance encompasses the structures, processes, and policies designed to ensure that AI systems operate ethically, safely, and in alignment with human values and legal requirements. This multifaceted approach addresses not just technical performance but also fairness, transparency, accountability, and the distribution of benefits across society. Effective AI governance requires a comprehensive understanding of four interconnected pillars that together form the foundation of responsible AI deployment.

The System

At the heart of AI governance lies the AI system itself. This includes everything from LLMs to ML algorithms that must be trained, tested, and integrated in a way that ensures they work together cohesively. Each component must be individually fine-tuned and then tested as part of the larger system to ensure seamless operation. The architecture, design choices, and technical implementation details all influence how the system will behave in production environments and what governance challenges it may present.

Tools

Managing and measuring the data that trains AI systems is no small feat. Tools are required to monitor everything from performance metrics like precision and recall to bias metrics such as demographic parity and equalized odds to drift metrics like population stability index. These tools help analyze the AI system's behavior and determine whether it meets the required ethical and functional standards. Without sophisticated monitoring and evaluation instruments, organizations cannot verify compliance with governance requirements or detect emergent problematic behaviors in deployed systems.

Data

Data is the lifeblood of AI systems. The more data available for training, the more accurate and effective the AI becomes. However, the data collected about an AI system during its development and deployment—including performance, bias, and drift metrics—provides insight into whether the system's behavior aligns with acceptable standards. Governance frameworks must address data provenance, quality, representativeness, and the ethical implications of how training data is collected, labeled, and used throughout the AI lifecycle.

Stakeholders

A diverse network of stakeholders is involved in AI governance, from engineers and data scientists to business leaders, bias detection specialists, regulators, and end-users. Each group plays a vital role in ensuring that the AI system is not only effective but also fair and transparent. Collaboration among these stakeholders is key to creating an AI ecosystem that serves everyone equitably. Governance structures must facilitate meaningful participation from all affected parties, particularly those who may be disproportionately impacted by AI systems yet traditionally excluded from technical decision-making processes.

These four elements—systems, tools, data, and stakeholders—form an integrated framework that must be considered holistically for effective AI governance. When properly aligned, they create accountability mechanisms that ensure AI technologies advance social welfare, respect human autonomy, and operate within ethical boundaries. As AI capabilities continue to advance rapidly, developing robust governance approaches becomes not just a technical challenge but a societal imperative with far-reaching implications for our collective future.

The Importance of AI Governance

AI governance is more than just a set of rules; it’s the framework that ensures AI operates within boundaries that protect and empower us. It establishes clear standards and processes to guide how AI is developed, deployed, and applied, ensuring that its use is transparent, fair, and trustworthy. Think of AI governance as the guardrails that keep this powerful technology from veering off course, preventing it from becoming a tool of harm rather than one of progress.

AI governance is the key to unlocking AI's potential while safeguarding against its risks. It allows us to trust that the AI systems we interact with—whether in healthcare, education, finance, or any other field—are designed with our best interests at heart. By setting these standards, we can ensure that AI doesn’t just function efficiently, but ethically, keeping human values at the forefront of its evolution.

The Risks of Ignoring AI Governance

Without AI governance, the risk of AI failure looms large, whether in our daily interactions, its implementation in institutions, or within government systems. Imagine a world where AI operates without oversight—where algorithms decide without accountability and where the lines between ethical use and exploitation blur dangerously. Without proper governance, this dystopian reality could very well become our future.

The issues raised in this chapter and throughout the book—hallucinations, AI safety, risks, and biases—underscore the urgent need for AI governance. Responsible AI is not just a lofty ideal; it’s a necessity to create an equitable AI-driven world. And at the heart of ensuring that responsibility is reinforced is robust AI governance.

International AI Initiatives and Partnerships

AI governance is a complex and multifaceted endeavor, demanding collaboration across borders and disciplines. It's not a task that can be accomplished in isolation; rather, it requires the concerted efforts of international AI initiatives and partnerships. While we've touched upon some aspects of AI governance throughout this book, the following mind map offers a visual representation of the core elements involved. These elements include Ethical AI, Risk Management, Regulatory Aspects, and Transparency—all of which are interconnected and crucial for effective governance.

Table 7.3 - AI Governance Mindmap
Ethical AI Regulatory Risk Management Transparency
- Fairness
- Transparency
- Accountability
- Privacy
- National Compliance
- International Laws
- Industry Specific
- Risk Identification
- Assessment
- Mitigation
- Explainability
- Model Interpretability
- User Understanding

Forging international AI initiatives and partnerships is a monumental task, requiring careful attention to multiple complex dimensions such as ethical regulations, risk management, transparency, and international cooperation. The key areas of concern include:

  • Ethical AI: Ensuring fairness, accountability, privacy, and transparency in AI systems.
  • Regulatory Compliance: Navigating national and international laws, industry-specific guidelines, and ensuring adherence to global standards.
  • Risk Management: Identifying, assessing, and mitigating risks, particularly related to bias, security, and privacy concerns.
  • Model Interpretability and Explainability: Making AI systems understandable to users and stakeholders, which is crucial for fostering trust and accountability.

While the United States currently leads in AI development and ethical governance, other nations and organizations are becoming increasingly involved. The effort to create global AI governance involves a variety of stakeholders, including governments, organizations, and international summits. While coordinating these elements is far from easy, shared understanding helps establish common standards to mitigate concerns around bias, privacy, security, and public trust. A range of entities—including governments, international summits, and cross-cultural organizations—are working to shape the global AI landscape.[83]

Several significant partnerships and alliances are emerging, each aiming to tackle these complex issues. For instance, the Global Partnership on Artificial Intelligence (GPAI) plays a key role in fostering international cooperation. Both India and the United States have joined forces to build a scientific consensus on the risks emerging from AI systems. India and the United States, in particular, have a joint interest in using AI to enhance cybersecurity. This was exemplified by the launch of the Defense Artificial Intelligence Dialogue in April 2022 to further cement their shared interest in leveraging AI for defense and security.[84]

Another example of cross-institutional collaboration is the memorandum of understanding signed in February 2024 between the University of Waterloo and Georgia Institute of Technology. This agreement underscores a commitment to strengthen academic and research ties, fostering joint initiatives in research, education, and innovation. Both universities are recognized globally for their impact, and this partnership represents an important step toward international AI cooperation in academia.[85]

Beyond academic partnerships, private sector collaborations are also advancing the AI agenda. Companies like Microsoft and Coca-Cola are experimenting with AI technologies, such as the Azure OpenAI Service and Microsoft Copilot, to develop innovative use cases across various business functions. These efforts aim to enhance workplace productivity for Coca-Cola’s 700,000 employees globally while aligning with the company’s values. Additionally, Microsoft is working to bring AI to underserved regions, expanding access to digital technology while ensuring adherence to global safety and security standards. As noted in a recent Microsoft press release, “This expanded collaboration will empower organizations of all sizes in new markets to harness the benefits of AI and the cloud while ensuring they are adopting AI that adheres to world-leading standards in safety and security."[86]

On a global scale, the Franco-Canadian GPAI and the OECD announced an “integrated partnership” in July 2024. This partnership involves government ministers from several countries and aims to combine the AI efforts of GPAI and OECD, extending their work to 44 countries, with a particular focus on low- and middle-income nations. Despite these positive developments, achieving global consensus on AI governance remains a significant challenge.[87]

China, one of the few serious competitors to the United States in AI development, took a significant step by forging an AI governance initiative with France in the summer of 2024. This move is seen as a direct challenge to U.S. leadership in AI governance, accelerating the race for global AI dominance. Experts suggest that the joint declaration will not only facilitate practical cooperation but also serve as a model for strengthening AI exchanges and partnerships between China and other European nations.[88]

As the international AI community continues to grapple with ethical, regulatory, and risk management issues, the path to cohesive global AI initiatives faces numerous obstacles. Addressing these challenges is crucial as we move forward to the next section.

Challenges in International AI Collaboration

A collage of four futuristic scenes showcasing AI integration in different industries. The top-left panel depicts an AI-controlled construction site with autonomous excavators and robotic assistants working on infrastructure projects. The top-right panel features a high-tech laboratory where a humanoid robot interacts with a scientist, alongside an autonomous research vehicle. The bottom-left panel highlights AI-driven transportation with self-driving taxis and a towering quadrupedal construction robot in an urban setting. The bottom-right panel shows two healthcare professionals using AI-powered diagnostic tools in a medical facility. Together, these images illustrate the widespread application of AI in automation, construction, healthcare, and transportation.
Figure 7.19 - AI in Industry and Infrastructure

Coordinating these elements is anything but simple. Who gets to decide, and what exactly they are deciding, is just as important as the technology itself. Is it the companies developing these AI models? Is it governments, or perhaps regulatory institutions?

Take, for instance, OpenAI's o1 and o1-mini models released in September 2024. The company proudly highlighted its enhanced reasoning capabilities, boasting that the new model solves math problems from the International Math Olympiad with an 83% success rate—compared to just 13% from its predecessors.[89] It performs at the level of PhD students on benchmarks in physics, chemistry, and biology, and achieves 89% accuracy in coding, excelling in Codeforces competitions.

These advancements, however, have raised significant concerns among experts. As AI's ability to reason at such high levels increases, so do the risks. "If OpenAI has indeed crossed into what they themselves describe as a 'medium risk' level for CBRN (chemical, biological, radiological, and nuclear) weapons, as they report, this underscores the urgent need for legislation like SB 1047 to protect the public," said AI scientist Yoshua Bengio in a statement to Newsweek. He continued, "The improvement of AI's reasoning abilities, especially when used to deceive, is particularly dangerous. We must implement regulatory measures like SB 1047 to ensure developers place public safety at the forefront of AI innovation."[90]

Highlights a growing recognition of the need for regulatory oversight, a proposed bill aimed at holding companies accountable for AI model malfunctions. According to this bill, the National Telecommunications and Information Administration (NTIA) must research and document artificial intelligence (AI) system accountability measures to ensure that the pursuit of AI advancement does not compromise public safety. Specifically, the NTIA is required to research, gather input from stakeholders, and report to Congress on methods (such as audits, certifications, and evaluations) that offer guarantees of incorporated accountibility measures within AI system. The demand for such bill underscores the local and global challenges posed by AI.[91]

To address these challenges, global AI governance must confront several complex issues: aligning AI development with diverse values, ensuring fair distribution of benefits, and mitigating risks that impact both current and future generations. The stakes are high, requiring a cohesive international framework that fosters collaboration and accountability.

AI transcends borders, presenting both tremendous opportunities and significant threats with global repercussions. Researchers advocate for the urgent need for international cooperation between researchers, organizations, and governing bodies..[92] However, numerous obstacles remain. One pressing issue is the unequal distribution of AI's benefits. While AI has the potential to revolutionize climate change mitigation, education, and healthcare, profit-driven corporations may prioritize financial gain over global welfare, exacerbating technological divides between affluent nations and marginalized communities.

Another critical challenge is managing global risks associated with AI, such as biases, system failures, and the potential for dangerous applications in military and surveillance contexts. The 2024 discussion paper by the Center of International Governance Innovation (CIGI) highlights that while recent global collaboration efforts are commendable, they fall short of addressing these issues comprehensively.[93]

A conceptual diagram illustrating the risks associated with artificial intelligence (AI). At the center is an orange silhouette of a human head with white circuit-like patterns, symbolizing the intersection of AI and human thought. The head is surrounded by three labeled circles, each representing a category of AI risk: AI Weaponization (green circle, top): Highlights the risks of using AI in harmful ways, encompassing both physical spaces (e.g., autonomous weapons) and virtual spaces (e.g., cyberattacks). System Failure (blue circle, right): Focuses on the dangers of AI systems malfunctioning, including "Dangerous Applications" that cause harm and "Loss of Control" when humans cannot manage AI behavior. Bias (purple circle, bottom): Emphasizes how AI systems can perpetuate or exacerbate issues like discrimination, inequality, and inequity due to biased algorithms or datasets. The diagram underscores the multifaceted risks posed by AI and the need for ethical oversight and robust safeguards in its development and application.
Figure 7.4 - AI Risk

AI weaponization involves the deliberate or inadvertent use of AI technologies in ways that can harm stakeholders, frequently through algorithmic malfunctions that compromise safety or obscure transparency. This danger is amplified by the rapid deployment of AI systems, which can fuel misinformation and disinformation, like AI-generated fake news or deepfakes, raising alarm about the potential for widespread harm. However, the risks of weaponization extend far beyond the mere spread of deceptive content. At its most dangerous, AI weaponization encompasses the development of advanced military technologies—autonomous drones programmed to identify and eliminate targets, or AI systems like Lavender deployed in conflict zones.

Control is becoming increasingly urgent: as AI systems advance, the potential for superintelligent AI to act autonomously—or be exploited by malicious actors—could result in catastrophic consequences. These risks include the application of system malfunctions to devastating scenarios in fields like biology and chemistry, or even in routine, everyday interactions.

AI systems often mirror the biases embedded in their training data, raising ethical concerns about fairness and representation. However, these challenges are only part of a much larger conversation about AI weaponization. The profound ethical implications of AI being used as a tool for harm—whether through misinformation, biased decision-making, or autonomous military actions—underscore the urgent need for stringent oversight and regulation to ensure AI is deployed responsibly.

As indicated by the CIGI, robust international cooperation is essential for developing mechanisms to monitor dangerous AI systems, identify those posing unacceptable risks, and, if necessary, prevent their development. The UN's 2024 resolution on safe, secure, and trustworthy AI, adopted unanimously by the General Assembly, outlines the benefits and risks of AI (United Nations General Assembly, 2024). Supporting this initiative, the UN Secretary-General has established a High-level Advisory Body on Artificial Intelligence to provide recommendations for globally coordinated AI governance.[94] Additionally, the UN Summit of the Future, scheduled for September 2024, aims to finalize a Global Digital Compact that will establish shared global values and principles for AI development.

International cooperation extends beyond the UN to include the G7, the OECD, the African Union, the European Union, and various bilateral and multilateral discussions inspired by the UK's AI Safety Summit.[95] [96]

But as the CIGI stresses, such initiatives, although important are not merely enough. To effectively manage these efforts, a global framework convention is essential. This convention should articulate a clear objective, such as "ensuring the development of beneficial, safe, and inclusive AI for the good of all humanity." Principles of cooperation, equity, inclusivity, effectiveness, and preparedness must guide it. Addressing the global challenges of AI requires unprecedented international agreement and collaboration. Without proactive and unified action, there is a risk of deepening existing divides and allowing AI’s dangers to overshadow its potential benefits.[97]

Key Takeaways

  • International collaboration is essential for setting consistent AI safety standards and ethical practices worldwide.
  • Organizations like the OECD and UNESCO play crucial roles in promoting safe and ethical AI development.
  • Shared guidelines and regulations help prevent the misuse of AI, promoting responsible and ethical AI use across borders.
  • While creating a unified international approach to AI safety is challenging, it is vital for addressing global concerns about AI.

Exercises

  • Global AI Standards Exploration: Research the AI safety guidelines of a specific country and compare them to international standards set by organizations like UNESCO.
  • AI Safety Proposal: Develop a proposal outlining steps your country could take to align with global AI safety standards.
  • Collaboration Case Study: Analyze a case where international collaboration has successfully promoted AI safety, discussing key takeaways for future efforts.

Chapter Wrap-Up

In this chapter, we explored the complex interplay between AI and privacy, focusing on both individual and societal concerns. Using the metaphor of the Panopticon, we examined the pervasive nature of AI surveillance and its implications for privacy. By emphasizing the importance of ethical design in data collection, we underscored the need for responsible data practices that prioritize transparency and user consent.

We also addressed AI's impact on employment, considering both its potential for job displacement and job enhancement. In particular, we highlighted the role AI can play in closing the digital divide by creating opportunities for more equitable access to work. The discussion then turned to the risks and challenges posed by AI, particularly weaponization, stressing the urgency of stakeholder collaboration to ensure AI systems are not only effective but also fair and sustainable.

Finally, in examining new approaches to fostering stakeholder collaboration, such as the global framework proposed by CIGI, we concluded that diversity and inclusivity are essential for driving innovative solutions. AI systems designed with a wide range of perspectives are better equipped to serve all users equitably, ultimately contributing to a more just and inclusive society.

Key Terms

  • AI Surveillance
  • Data Anonymization
  • Data Privacy
  • Digital Divide
  • Identifiability
  • Job Displacement
  • Predictive Policing
  • Soft Skills

Chapter Exercises

Imagine you are developing an AI application that collects user data to improve personalization. What measures would you implement to ensure data privacy, such as data anonymization or adherence to GDPR guidelines? Discuss how these measures balance user privacy with the need for data-driven insights.

Choose a sector, such as retail, healthcare, or manufacturing, and research how AI is currently transforming job roles within it. Write a report or create a presentation on the roles most impacted by automation, new opportunities created, and strategies that organizations in this sector are using to reskill and upskill their workforce.

Design and conduct a survey to assess public concerns about privacy in AI technologies, focusing on areas like data collection, transparency, and user control. Analyze your results to identify key trends and propose solutions that companies or governments could implement to address these privacy concerns.

Develop a proposal for an international initiative to address a specific AI-related safety or ethical concern (e.g., algorithmic bias, facial recognition regulations). Outline the project's goals, the roles of participating countries, and the benefits of international collaboration on this issue. Include potential challenges and strategies for overcoming them.

Research an existing reskilling or upskilling program designed to help workers adapt to AI-driven changes in the job market. Evaluate the program’s structure, effectiveness, and areas for improvement. Discuss how similar programs could be implemented more widely to prepare the workforce for an AI-enhanced future.

Real-World Case Study

 Italy's Landmark GDPR Fine Against OpenAI

In late 2024, Italy's Data Protection Authority issued a groundbreaking €15 million fine to OpenAI, marking the first generative AI-related case brought under GDPR. The decision highlighted the growing tension between AI development and data privacy regulations.

The investigation revealed multiple GDPR violations in OpenAI's operation of ChatGPT. The primary issue centered on the company's use of personal data to train its AI models without establishing a proper legal basis - a fundamental requirement under GDPR. Additionally, Garante found that OpenAI failed to implement adequate age verification systems to protect users under 13 from inappropriate AI-generated content and did not properly notify authorities about a data breach that occurred in March 2023.

Beyond the monetary penalty, the ruling included a unique remedial measure: OpenAI was required to conduct a six-month public education campaign in Italian media to inform users about ChatGPT's data collection practices and individuals' rights under GDPR. This case was particularly significant as it followed Italy's temporary ban of ChatGPT in March 2023, demonstrating escalating regulatory oversight of AI technologies in Europe.

The case raises several critical questions about AI development and privacy regulation:

How can AI companies balance the need for vast training data with GDPR's strict requirements for legal basis and consent?

What constitutes adequate transparency in AI systems, particularly regarding complex processes like model training?

How might this precedent-setting fine influence other EU regulators' approach to AI oversight?

What are the implications for global AI development when companies face different privacy standards across jurisdictions?

This case illustrates the growing challenges AI companies face in navigating complex privacy regulations while pursuing technological innovation. OpenAI's experience demonstrates that even industry leaders must adapt their practices to meet regional privacy standards, potentially setting new benchmarks for responsible AI development worldwide.

BMW's AI Integration and Workforce Evolution

In 2018, BMW Group began a significant transformation of its production processes by introducing AI applications across its manufacturing plants, with its Dingolfing facility serving as a primary testing ground.[98] Rather than replacing workers, BMW implemented a strategy that focused on using AI to enhance human capabilities while transitioning workers to new roles.

The company deployed AI in several key areas: automated image recognition for quality control, logistics robots for material transport, and sophisticated production management systems. In the press shop, for example, AI systems were implemented to distinguish between actual defects and pseudo-defects in car body components, a task that previously required extensive manual inspection. The company emphasized that AI implementation was designed to relieve workers of repetitive tasks while maintaining high-quality standards.

What made BMW's approach notable was its focus on employee involvement in the AI transition. Workers were empowered to identify areas where AI applications could improve quality and efficiency, and the company deliberately kept the setup and implementation of AI applications simple enough that no advanced IT proficiency was required. By 2024, the company had expanded its AI implementation to include humanoid robots from Figure AI, representing a new phase in human-robot collaboration.[99] [100]

The case raises several critical questions:

Questions:

How can manufacturers effectively balance automation with workforce retention and development?

What role should employees play in identifying and implementing AI solutions in their workplaces?

How can companies ensure that AI implementation enhances rather than replaces human capabilities?

What are the implications of introducing humanoid robots alongside human workers in manufacturing settings?

This case demonstrates how systematic AI integration can enhance manufacturing processes while maintaining workforce stability through strategic role transitions and employee involvement in technological implementation.

End-of-Chapter Assessment

Discussion Questions

How can AI pose privacy risks, and what measures can be taken to protect user privacy?

In what ways can AI impact employment positively and negatively across different job sectors?

What are some ethical considerations for AI in the workplace, and how can businesses address potential job displacement?

How does international collaboration contribute to AI safety, and why is it crucial for regulating AI technologies?

What are the challenges and benefits of creating unified global standards for AI safety?

Multiple Choice Questions

1. Which of the following is a primary privacy concern associated with AI?

A) Reduced internet speed
B) Excessive data collection
C) Lower costs for businesses
D) Improved customer service

2. What type of regulation is GDPR most associated with?

A) Environmental protection
B) Data privacy
C) Employment standards
D) International trade

3. What is a primary benefit of AI in the workforce?

A) Job displacement
B) Enhanced automation of repetitive tasks
C) Increased unemployment
D) Reduced productivity

4. Which skills are likely to become more valuable as AI transforms job markets?

A) Physical labor skills
B) Basic computational skills
C) Critical thinking and creativity
D) Simple clerical skills

5. Why is international collaboration essential for AI safety?

A) It promotes competition between countries
B) It helps set global standards and prevent misuse of AI
C) It encourages companies to disregard regulations
D) It reduces the need for innovation

6. Which organization is known for setting international guidelines for AI ethics?

A) UNESCO
B) FDA
C) IRS
D) NASA

7. What is one of the main challenges in establishing unified AI safety standards globally?

A) Lack of interest among countries
B) Cultural and regulatory differences
C) Absence of AI applications in many countries
D) High agreement on safety practices

8. Which of the following is an effective method for enhancing data privacy in AI systems?

A) Reducing transparency
B) Data anonymization
C) Limiting user control over data
D) Ignoring data protection laws

9. What impact does AI-driven job displacement have on the workforce?

A) It increases overall job availability
B) It reduces the need for skilled labor
C) It can lead to higher unemployment if no reskilling efforts are made
D) It has no impact on existing jobs

10. What is one benefit of reskilling programs in response to AI's impact on employment?

A) They decrease workforce adaptability
B) They prepare workers for new roles in an AI-driven market
C) They increase job displacement
D) They discourage employee growth

True or False

AI poses no risks to personal privacy.

GDPR is a data privacy regulation implemented in the United States.

AI can lead to both job creation and job displacement.

Soft skills like creativity and critical thinking are less valuable in an AI-driven workforce.

International collaboration is unnecessary for AI safety.

Data anonymization is a technique used to protect user privacy in AI systems.

Countries face challenges in creating unified AI safety standards due to regulatory differences.

The OECD is an organization involved in setting AI safety guidelines.

AI-driven automation is expected to decrease the need for reskilling in the workforce.

AI can impact employment positively by taking over repetitive tasks and increasing efficiency.

Answer Key

Discussion Questions

How can AI pose privacy risks, and what measures can be taken to protect user privacy?
Example Answer: AI can invade privacy by collecting, storing, and analyzing large amounts of personal data. Measures like data anonymization, transparency in data practices, and regulatory frameworks like GDPR can help protect user privacy.

In what ways can AI impact employment positively and negatively across different job sectors?
Example Answer: AI can positively impact employment by creating new roles in tech and automation sectors, but it may negatively impact traditional roles that can be automated, leading to potential job displacement.

What are some ethical considerations for AI in the workplace, and how can businesses address potential job displacement?
Example Answer: Ethical considerations include the fair treatment of displaced workers and equitable job opportunities. Businesses can support reskilling and upskilling programs to help workers transition to new roles.

How does international collaboration contribute to AI safety, and why is it crucial for regulating AI technologies?
Example Answer: International collaboration promotes shared standards, preventing AI misuse and fostering global responsibility. It’s essential because AI impacts society universally and requires consistent regulations.

What are the challenges and benefits of creating unified global standards for AI safety?
Example Answer: Challenges include regulatory and cultural differences, while benefits include enhanced trust, reduced misuse of AI, and a collaborative approach to innovation and safety.

Multiple Choice

Which of the following is a primary privacy concern associated with AI?
Answer: B. Excessive data collection

What type of regulation is GDPR most associated with?
Answer: B. Data privacy

What is a primary benefit of AI in the workforce?
Answer: B. Enhanced automation of repetitive tasks

Which skills are likely to become more valuable as AI transforms job markets?
Answer: C. Critical thinking and creativity

Why is international collaboration essential for AI safety?
Answer: B. It helps set global standards and prevent misuse of AI

Which organization is known for setting international guidelines for AI ethics?
Answer: A. UNESCO

What is one of the main challenges in establishing unified AI safety standards globally?
Answer: B. Cultural and regulatory differences

Which of the following is an effective method for enhancing data privacy in AI systems?
Answer: B. Data anonymization

What impact does AI-driven job displacement have on the workforce?
Answer: C. It can lead to higher unemployment if no reskilling efforts are made

What is one benefit of reskilling programs in response to AI's impact on employment?
Answer: B. They prepare workers for new roles in an AI-driven market

True or False

AI poses no risks to personal privacy.
Answer: False – AI can pose significant privacy risks through data collection and storage practices.

GDPR is a data privacy regulation implemented in the United States.
Answer: False – GDPR is implemented in the European Union.

AI can lead to both job creation and job displacement.
Answer: True – AI creates new jobs in tech-related fields while potentially displacing others.

Soft skills like creativity and critical thinking are less valuable in an AI-driven workforce.
Answer: False – Soft skills like creativity and critical thinking are more valuable as they are less likely to be automated.

International collaboration is unnecessary for AI safety.
Answer: False – International collaboration is crucial for setting consistent safety standards in AI.

Data anonymization is a technique used to protect user privacy in AI systems.
Answer: True – Data anonymization helps reduce privacy risks by removing personal identifiers.

Countries face challenges in creating unified AI safety standards due to regulatory differences.
Answer: True – Differences in cultural and regulatory practices make it challenging to establish unified standards.

The OECD is an organization involved in setting AI safety guidelines.
Answer: True – The OECD is one of the organizations that promote AI safety and ethical guidelines.

AI-driven automation is expected to decrease the need for reskilling in the workforce.
Answer: False – Reskilling is essential to help workers adapt to changing job requirements due to AI.

AI can impact employment positively by taking over repetitive tasks and increasing efficiency.
Answer: True – AI can automate repetitive tasks, allowing workers to focus on more complex, creative tasks.

References


  1. Güven, Ç., Alishahi, A., Brighton, H., Nápoles, G., Olier, J. S., Šafář, M., Postma, E., Shterionov, D., De Sisto, M., & Vanmassenhove, E. (2025). AI in support of diversity and inclusion. ArXiv. https://arxiv.org/abs/2501.09534
  2. Ma, J. (2024). The impact of AI bias on social justice: challenges and solutions. Journal of Computing and Electronic Information Management, 15(3), 75–78. https://doi.org/10.54097/wz9v0f43
  3. Pols, A., Bałazińska, E., & Lubowicka, K. (2018, September 21). OECD guidelines: 8 privacy principles to live by. Piwik PRO. https://piwik.pro/blog/oecd-guidelines-8-privacy-principles-to-live-by/
  4. Worley, A., & Nguyen, T. (2024, February). Navigating the intersection of artificial intelligence, privacy, and data protection. THINKBRG.COM. https://media.thinkbrg.com/wp-content/uploads/2024/02/27120840/BRG-AI_Privacy_Data_Feb2024.pdf; para. 4.
  5. Campolo, A., Sanfilippo, M., Whittaker, M., & Crawford, K. (2017). AI Now 2017 Report. AI Now Institute. https://ainowinstitute.org/AI_Now_2017_Report.pdf
  6. Stone, P., Brooks, R., Brynjolfsson, E., Calo, R., Etzioni, O., Hager, G., Hirschberg, J., Kalyanakrishnan, S., Kamar, E., Kraus, S., Leyton-Brown, K., Parkes, D., Press, W., Saxenian, A., Shah, J., Tambe, M., & Teller, A. (2016). Artificial intelligence and life in 2030. One Hundred Year Study on Artificial Intelligence: Report of the 2015-2016 Study Panel. Stanford University. http://ai100.stanford.edu/2016-report
  7. Office of the Victorian Information Commissioner. (n.d.). Artificial intelligence and privacy – Issues and challenges. https://ovic.vic.gov.au/privacy/resources-for-organisations/artificial-intelligence-and-privacy-issues-and-challenges/
  8. Locher, J. (2024, August 1). IOC leader says 'hate speech' directed at Imane Khelif and Lin Yu-Ting at Olympics is unacceptable. Associated Press. https://www.kvue.com/article/sports/olympics/imane-khelif-lin-yu-ting-olympics-gender-scrutiny-ioc-president/507-d3a09016-2e69-4c9f-ab87-864fcbf1e0fc; para. 26.
  9. Locher, J. (2024, August 1). IOC leader says 'hate speech' directed at Imane Khelif and Lin Yu-Ting at Olympics is unacceptable. Associated Press. https://www.kvue.com/article/sports/olympics/imane-khelif-lin-yu-ting-olympics-gender-scrutiny-ioc-president/507-d3a09016-2e69-4c9f-ab87-864fcbf1e0fc; para. 10.
  10. Myth Detector. (2024, September 19). AI-generated photo circulates to claim boxer Imane Khelif is a transgender. Myth Detector. https://mythdetector.com/en/ai-generated-photo-circulates-to-claim-boxer-imane-khelif-is-a-transgender/
  11. Norberg, P. A., Horne, D. R., & Horne, D. A. (2007). The privacy paradox: Personal information disclosure intentions versus behaviors. Journal of Consumer Affairs, 41(1), 100–126. https://doi.org/10.1111/j.1745-6606.2006.00070.x
  12. Peacock, S. (2014). How web tracking changes user agency in the age of Big Data: The used user. Big Data and Society, 1(2). https://doi.org/10.1177/2053951714564228
  13. Office of the Victorian Information Commissioner. (n.d.). Artificial intelligence and privacy – Issues and challenges. https://ovic.vic.gov.au/privacy/resources-for-organisations/artificial-intelligence-and-privacy-issues-and-challenges/
  14. Walsh, T. (2017). It’s alive! Artificial intelligence from the logic piano to killer robots. Trobe University Press.
  15. Racine, E. E. (2023). The far-reaching implications of China’s AI-powered surveillance state post-COVID-19. Surveillance & Society, 21(3), 269–275. https://doi.org/10.24908/ss.v21i3.16111
  16. Wakefield, J. (2021, May 25). AI emotion-detection software tested on Uyghurs. BBC News. https://www.bbc.com/news/technology-57101248
  17. Human Rights Watch. (2019, May 1). China’s algorithms of repression: Reverse engineering a Xinjiang police mass surveillance app. https://www.hrw.org/report/2019/05/01/chinas-algorithms-repression/reverse-engineering-xinjiang-police-mass
  18. Foucault, M. (1977). Discipline and punish: The birth of the prison (A. Sheridan, Trans.). Vintage Books. (Original work published 1975).
  19. Foucault, M. (1977). Discipline and punish: The birth of the prison (A. Sheridan, Trans.). Vintage Books. (Original work published 1975; p. 205).
  20. Pfau, M. (2024, February 2). Artificial intelligence: The new eyes of surveillance. Forbes. https://www.forbes.com/sites/forbestechcouncil/2024/02/02/artificial-intelligence-the-new-eyes-of-surveillance/
  21. Lyon, D. (1994). The electronic eye: The rise of surveillance society. University of Minnesota Press. https://www.jstor.org/stable/10.5749/j.ctttsqw8
  22. Pfleeger, S. (2014). The eyes have it: Surveillance and how it evolved. IEEE Security & Privacy, 12(2), 74–79. https://doi.org/10.1109/MSP.2014.80
  23. Olvera, A. (2024, June 7). How AI surveillance threatens democracy everywhere. Bulletin of the Atomic Scientists. https://thebulletin.org/2024/06/how-ai-surveillance-threatens-democracy-everywhere
  24. Feldstein, S. (2019, September 17). The global expansion of AI surveillance. Carnegie Endowment for International Peace. https://carnegieendowment.org/research/2019/09/the-global-expansion-of-ai-surveillance
  25. Beraja, M., Kao, A., Yang, D. Y., & Yuchtman, N. (2023). AI-tocracy. The Quarterly Journal of Economics, 138(3), 1349–1402. https://doi.org/10.1093/qje/qjad012
  26. Olvera, A. (2024, June 7). How AI surveillance threatens democracy everywhere. Bulletin of the Atomic Scientists. https://thebulletin.org/2024/06/how-ai-surveillance-threatens-democracy-everywhere/
  27. Gunkel, D. J. (Ed.). (2024). Handbook on the ethics of artificial intelligence. Northampton, MA: Edward Elgar Publishing.
  28. Li, P., Yang, J., Islam, M. A., & Ren, S. (2023). Making AI less "thirsty": Uncovering and addressing the secret water footprint of AI models. ArXiv. https://arxiv.org/abs/2304.03271
  29. Elmoudden, S., & Wrench, J. S. (2024). The future is already here: From GAI to AGI. In S. Elmoudden & J. S. Wrench (Eds.), The role of generative AI in the communication classroom (pp. 316–336). IGI Global.
  30. Tacheva, Z., & Ramasubramanian, S. (2023). AI empire: unraveling the interlocking systems of oppression in generative AI's global order. Big Data & Society, 10(2). https://doi.org/10.1177/20539517231219241
  31. Laney, D. (2001, February 6). 3D data management: Controlling data volume, velocity, and variety (File 949). META Group.
  32. National Institute of Standards and Technology (NIST). (2015). NIST Big Data Interoperability Framework (vol. 1 - definitions). https://doi.org/10.6028/NIST.SP.1500-1
  33. Kerry, C. F. (2020, February 10). Protecting privacy in an AI-driven world. Brookings Institutue. https://www.brookings.edu/research/protecting-privacy-in-an-ai-driven-world/
  34. Skolnik, S. (2024, August 5). OpenAI was sued over using YouTube videos without creators’ consent. Bloomberg Law. https://news.bloomberglaw.com/litigation/openai-sued-over-using-youtube-videos-without-creators-consent
  35. Sundara Rajan, M. T. (2024, February 29). Is generative AI fair use of copyright works? NYT v. OpenAI. Kluwer Copyright Blog. https://tinyurl.com/bdf7wnpr
  36. Health Insurance Portability and Accountability Act. Pub. L. No. 104–191, § 264, 110 Stat.1936.
  37. Murdoch, B. (2021). Privacy and artificial intelligence: Challenges for protecting health information in a new era. BMC Medical Ethics, 22(1). https://doi.org/10.1186/s12910-021-00687-3
  38. Gonçalves, A., Ray, P., Soper, B., Stevens, J., Coyle, L., & Sales, A. P. (2020). Generation and evaluation of synthetic patient data. BMC Medical Research Methodology, 20(1). https://doi.org/10.1186/s12874-020-00977-1
  39. [footnote]Department of Health and Human Services. (2020). Guidance for regulation of artificial intelligence applications (OMB M-21-06). https://www.hhs.gov/sites/default/files/department-of-health-and-human-services-omb-m-21-06.pdf
  40. Office for Civil Rights. (2024, December 27). HIPAA security rule notice of proposed rulemaking to strengthen cybersecurity for electronic protected health information. U.S. Department of Health and Human Services. https://www.hhs.gov/hipaa/for-professionals/security/hipaa-security-rule-nprm/factsheet/index.html
  41. Antunes, R. S., Costa, C. A. d., Küderle, A., Yari, I. A., & Eskofier, B. M. (2022). Federated learning for healthcare: Systematic review and architecture proposal. ACM Transactions on Intelligent Systems and Technology, 13(4), 1–23. https://doi.org/10.1145/3501813
  42. IBM Corporation. (2024). Cost of a data breach report 2024. IBM.
  43. Mayover, T. (2024, October 2). When AI technology and HIPAA collide. The HIPAA Journal. https://www.hipaajournal.com/when-ai-technology-and-hipaa-collide/
  44. Klimczuk-Kochańska, M., & Klimczuk, A. (2015). Technological unemployment. In M. Odekon (Ed.), The SAGE encyclopedia of world poverty (2nd ed., pp. 1510–1511). SAGE Publications. https://doi.org/10.4135/9781483345727.n783
  45. Frey, C. B., & Osborne, M. A. (2017). The future of employment: how susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254–280. https://doi.org/10.1016/j.techfore.2016.08.019
  46. McKinsey Global Institute. (2017). Jobs lost, jobs gained: Workforce transitions in a time of automation. McKinsey & Company.
  47. World Economic Forum. (2020, October 20). The future of jobs report 2020. http://www3.weforum.org/docs/WEF_Future_of_Jobs_2020.pdf
  48. World Economic Forum. (2023, April 30). The future of jobs report 2023. https://www3.weforum.org/docs/WEF_Future_of_Jobs_2023.pdf
  49. World Economic Forum. (2025, January 7). The future of jobs report 2025. https://reports.weforum.org/docs/WEF_Future_of_Jobs_Report_2025.pdf
  50. Kaufman, M. (2025, February 27). It's America's fastest-growing job – thanks to ChatGPT. CNN. https://www.cnn.com/2025/02/27/business/chatgpt-ai-jobs-surge/index.html
  51. Mok, A. (2023, May 3). AI won't take your job, but someone who uses it might. Business Insider. https://www.businessinsider.com/ai-wont-take-your-job-someone-who-uses-it-might-2023-5
  52. Gunkel, D. J. (Ed.). (2024). Handbook on the ethics of artificial intelligence. Edward Elgar Publishing
  53. Würzburg, T. (2023, June 19). AI alters your brain: Unveiling the impact of technology on your cognitive functioning. The Neural Network. LinkedIn. https://www.linkedin.com/pulse/brain-power-vs-ai-how-balance-technology-use-health-tom-w%C3%BCrzburg/
  54. McCoy, J. (2024, May 11). 100M jobs gone because of AI. What will we do? [Video]. YouTube. https://www.youtube.com/watch?v=xbNDNEBuTgs
  55. CBS Evening News. (2024, January 28). Automated AI restaurant opens in California [Video]. YouTube. https://www.youtube.com/watch?v=zyUekx9NZ18
  56. Kasár, R. (2024, March 28). Is AI poised to redefine the barista role? Allegra: World Coffee Portal. https://www.worldcoffeeportal.com/Latest/InsightAnalysis/2024/March/Rise-of-the-machines-Is-AI-poised-to-redefine-the
  57. Goldman Sachs. (2023, April 5). Generative AI could raise global GDP by 7%. https://www.goldmansachs.com/insights/articles/generative-ai-could-raise-global-gdp-by-7-percent.html
  58. Tiwari, R. (2023). The impact of AI and machine learning on job displacement and employment opportunities. International Journal of Scientific Research and Engineering Management, 7(1). https://doi.org/10.55041/ijsrem17506
  59. Johnson, A. (2023, March 30). Which jobs will AI replace? These 4 industries will be heavily impacted. Forbes. https://tinyurl.com/57jwxhc4
  60. World Economic Forum. (2020, October 20). The future of jobs report 2020. http://www3.weforum.org/docs/WEF_Future_of_Jobs_2020.pdf
  61. Davey, J. (2023, February 12). 65% of retail jobs could be automated by 2025. Freethink. https://www.freethink.com/robots-ai/retail-artificial-intelligence-robots
  62. PricewaterhouseCoopers. (2018). Will robots really steal our jobs? An international analysis of the potential long term impact of automation. https://www.pwc.com/hu/hu/kiadvanyok/assets/pdf/impact_of_automation_on_jobs.pdf
  63. Talmage-Rostron, M. (2024, January 10). How will artificial intelligence affect jobs 2024-2030? Nexford University. https://www.nexford.edu/insights/how-will-ai-affect-jobs
  64. Ellingrud, K., Sanghvi, S., Dandona, G. S., Madgavkar, A., Chui, M., White, O., & Hasebe, P. (2023, July 26). Generative AI and the future of work in America. McKinsey Global Institute. https://www.mckinsey.com/mgi/our-research/generative-ai-and-the-future-of-work-in-america
  65. Aimesoft Multimodal AI Services, Solutions & Consulting. (n.d.). AimeReception: Virtual receptionist, AI avatar, virtual human based on multimodal AI logo. https://www.aimesoft.com/aimereception.html
  66. Hinton, G. (2016, November 24). On radiology [Video]. YouTube. https://www.youtube.com/watch?v=2HMPRXstSvQ
  67. GE Healthcare. (2021, September 2). Unleashing the promise of artificial intelligence in radiology. GE Healthcare. https://www.gehealthcare.com/insights/article/unleashing-the-promise-of-artificial-intelligence-in-radiology
  68. Wood, T. (2024, March 15). Will AI replace doctors, lawyers, writers, engineers, or radiologists? Fast Data Science. https://fastdatascience.com/ai-in-healthcare/ai-replace-radiologists-doctors-lawyers-writers-engineers/
  69. CGTN BIZ. (2024, May 22). AI robots designed to support the elderly surge in China [Video]. YouTube. https://www.youtube.com/watch?v=EhkE4nyalwY
  70. Weimann-Sandig, N. (2024). How the use of AI is changing the role of educators at universities – And why this is by no means a bad thing. In INTED2024 Proceedings 18th International Technology, Education and Development Conference, Valencia, Spain. IATED. (pp. 394–400). https://doi.org/10.21125/inted.2024.0155
  71. The Museum of Modern Art. (2023, March 15). AI art: How artists are using and confronting machine learning ; How to see like a machine [Video]. YouTube. https://www.youtube.com/watch?v=G2XdZIC3AM8
  72. The Museum of Modern Art. (2023, March 15). AI art: How artists are using and confronting machine learning; How to see like a machine [Video]. YouTube. https://www.youtube.com/watch?v=G2XdZIC3AM8; paras. 4 & 5.
  73. Frey, C. B. and Osborne, M. A. (2017). The future of employment: how susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254–280. https://doi.org/10.1016/j.techfore.2016.08.019
  74. World Economic Forum. (2020). The future of jobs report 2020. https://www.weforum.org/reports/the-future-of-jobs-report-2020
  75. Topol, E. (2019). Deep medicine: How artificial intelligence can make healthcare human again. Basic Books.
  76. Van Durme, Y., Scoble-Williams, N., Eaton, K., Kirby, L., Griffiths, M., Poynton, S., Mallon, D., & Forsythe, J. (2023, January 9). Deloitte 2023 Global human capital trends: New fundamentals for a boundaryless world. Deloitte. https://www2.deloitte.com/us/en/insights/focus/human-capital-trends/2023/future-of-workforce-management.html
  77. McKinsey Global Institute. (2023). Jobs lost, jobs gained: Workforce transitions in a time of automation. McKinsey & Company.
  78. Aoun, J. E. (2024). Robot-proof: Higher education in the age of artificial intelligence. MIT Press.
  79. Wolfe, M. (2024, May 16). How AI is changing the world | A chat with Matt Wolfe [Video]. YouTube. https://www.youtube.com/watch?v=NGgY-YJRJAk
  80. Bilyeu, T. (2023, July 18). The 3-year AI reset: How to get ahead while others lose their jobs (Prepare now) | Emad Mostaque [Video]. YouTube. https://www.youtube.com/watch?v=Se91Pn3xxSs
  81. France, E. (2024, June 4). Using AI to decode dog vocalizations. University of Michigan News. https://news.umich.edu/using-ai-to-decode-dog-vocalizations/
  82. Credo AI. (2022, November 10). What is AI governance and why should you care? https://www.credo.ai/blog/cutting-through-the-noise-what-is-ai-governance
  83. Georgia Tech. (2024, February 12). Georgia Tech, University of Waterloo forge partnership to advance AI initiatives. Georgia Institute of Technology News. https://news.gatech.edu/news/2024/02/12/georgia-tech-university-waterloo-forge-partnership-advance-ai-initiatives
  84. Mohanty, A., & Singh, A. (2024, February 20). Advancing the India-U.S. partnership on AI. Carnegie Endowment for International Peace. https://carnegieendowment.org/posts/2024/02/advancing-the-india-us-partnership-on-ai?lang=en
  85. Martin, B. (2024, February 12). Georgia Tech, University of Waterloo forge partnership to advance AI initiatives. Georgia Tech News Center. https://tinyurl.com/3fv68t7v
  86. Microsoft Media Relations. (2024, April 15). Microsoft invests $1.5 billion in Abu Dhabi’s G42 to accelerate AI development and global expansion [Press release]. https://tinyurl.com/2mv5eb8e; para. 2.
  87. Hudson, R. L. (2024, July 4). Two global AI initiatives merge, aim to involve more developing countries in AI policy debate. Science Business. https://sciencebusiness.net/news/ai/two-global-ai-initiatives-merge-aim-involve-more-developing-countries-ai-policy-debate
  88. Liu, C., & Zhang, W. (2024, May 7). China, France release joint declaration on AI governance: Serves as model for enhancing partnerships between China and other European countries: Experts. Global Times. https://www.globaltimes.cn/page/202405/1311822.shtml
  89. OpenAI. (2024, September 12). Introducing OpenAI o1-preview. https://openai.com/index/introducing-openai-o1-preview/
  90. Boran, M. (2024, September 13). OpenAI o1 model warning issued by scientist: 'Particularly dangerous.' Newsweek. https://www.newsweek.com/openai-advanced-gpt-model-potential-risks-need-regulation-experts-1953311; para. 7.
  91. U.S. Congress. (2023). H.R.3369 - Artificial Intelligence Accountability Act, 118th Congress (2023-2024). Congress.gov. https://www.congress.gov/bill/118th-congress/house-bill/3369
  92. Lukacovic, M. N., & Sellnow-Richmond, D. D. (2024). Facing mega-crises in the global era of AI: The role of communication education. In S. Elmoudden & J. S. Wrench (Eds.), The role of generative AI in the communication classroom (pp. 146–166). IGI Global. https://doi.org/10.4018/979-8-3693-0831-8.ch008
  93. Araya, D., & He, A. (2024, December 4). United States-China multilateralism in the age of military AI (CIGI Paper No. 309). Center for International Governance Innovation. https://www.cigionline.org/documents/2885/Araya_and_Alex_He.pdf
  94. Advisory Body on Artificial Intelligence (2023). Interim Report: Governing AI for Humanity. United Nations. www.un.org/sites/un2.un.org/files/un_ai_advisory_body_governing_ai_for_humanity_ interim_report.pdf.
  95. African Union (June 17, 2024). African ministers adopt landmark continental artificial intelligence strategy, African Digital Compact to drive Africa’s development and inclusive growth [Press release]. https://tinyurl.com/3kkdub63
  96. Cass-Beggs, D., Clare, S., Dimowo, D., & Kara, Z. (2024). Framework convention on global AI challenges: Accelerating international cooperation to ensure beneficial, safe and inclusive AI (CIGI Discussion Paper). Centre for International Governance Innovation. https://www.cigionline.org/static/documents/AI-challenges.pdf
  97. Cass-Beggs, D., Clare, S., Dimowo, D., & Kara, Z. (2024). Framework convention on global AI challenges: Accelerating international cooperation to ensure beneficial, safe and inclusive AI (CIGI Discussion Paper). Centre for International Governance Innovation. https://www.cigionline.org/static/documents/AI-challenges.pdf
  98. BMW Group. (2019, July 15). Fast, efficient, reliable: Artificial intelligence in BMW Group production [Press release]. https://tinyurl.com/5xxekcp9
  99. BMW Group. (2024, November 25). High-tech in production: BMW Group enables automated driving for new vehicles [Press release]. https://tinyurl.com/bde9ypwc
  100. Hughes, L. (2024, July 8). Humanoid robots work the BMW factory floor. IoT World Today. https://www.iotworldtoday.com/robotics/humanoid-robots-work-the-bmw-factory-floor
definition

License

Icon for the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

The Future is Now: Empowering Society Through AI Literacy Copyright © by Jason S. Wrench and Sanae Elmoudden is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, except where otherwise noted.