Surge in AI Ethics Program Enrolment Sets the Stage for Students and Teachers to Succeed in the Battle for Responsible AI Development

 

By James Myers

The public reach and influential power of artificial intelligence is rapidly expanding, boosted by the November 2022 headline-making launch of ChatGPT and its large language model technology. AI’s swift introduction to the world stage has, however, unleashed a host of new ethical issues. It has also amplified some thorny problems that have become prominent over the 25-year history of our digital era, which erupted with widespread adoption of the internet at the end of the 1990s. 

As the power and influence of AI entrenches around the globe, regulators have been unable or unwilling to keep pace with mounting and widely publicized ethical concerns. The consequences for some AI users have been severe. Two prominent examples among a much larger number are the tragic ChatGPT-encouraged suicide of 16-year-old Adam Raine and the proliferation of a condition called AI psychosis. (For more on these issues, read The Quantum Record’s October 2025 feature Emerging Risks of AI Chatbots Include Suicide and “AI Psychosis,” Particularly for Vulnerable Youth.) 

Although regulatory defences in the battle for ethical AI are proving inadequate or ineffective, the rapidly increasing number of computer science students enrolled in ethical AI courses gives hope for victory to proponents of ethical AI development. 

Programs in AI ethics, like AI itself, are a relatively new emergence in the academic landscape, and so there are few systematic studies of their different approaches and scope. A simple internet search on the subject of AI ethics courses, however, reveals a large offering in all areas of the globe. It’s a clear sign that supply is responding to demand, and that demand is rapidly rising. 

Increasing global interest in AI ethics training will help to equip teachers and students with the skills necessary for designing ethical systems, to teach the next generations about the importance of ethics in AI, and to encourage the commercial proliferation of ethical AI. If ethical AI can prove itself more economically sustainable than the alternative, commercial interests should naturally begin to align within ethical boundaries. 

Here are a few examples of the many courses and programs in AI ethics available globally.

Launched in 2020, the University of Toronto’s Embedded Ethics Education Initiative (E3I) aims to help students “recognize the broader ramifications of the technology they’re developing on diverse stakeholders, and to avoid or mitigate any negative impact,” according to computer science Professor Sheila McIlraith. McIlraith is a co-creator the E3I program, in collaboration with the university’s department of philosophy. 

In the 2024 academic year, modules from the E3I program were incorporated in computer science courses attended by 8,000 students and, beginning in 2025, each computer science undergraduate will encounter E3I modules in at least one and sometimes two or more courses each year. 

E3I students are taught, for example, how a particular data structure can lead to development of an unauthorized contact tracing system. In such cases, students assess the ethical issues of data collection and consequences to the privacy of users. Course modules are available online, with subjects including Addiction and Computer Game Design, Anthropomorphization in Natural Language Processing, and Bias in Image Understanding Algorithms.

In another example that made its debut in the fall of 2024, MIT’s Ethics of Computing course is co-taught by Professor of Computing Armando Solar-Lezama and Philosophy Professor Brad Skow. Students enrolled in the course at the Cambridge, Massachusetts-based university receive alternating lectures from Solar-Lezama and Skow. The course begins with questions about societal-level AI risks, and as part of their coursework students examine the ethics of AI applications like COMPAS, whose controversial algorithms predict the likelihood that someone accused of crime would re-offend. In 2016, a ProPublica investigation of 10,000 prisoners subjected to profiling by COMPAS “found that black defendants were far more likely than white defendants to be incorrectly judged to be at a higher risk of recidivism, while white defendants were more likely than black defendants to be incorrectly flagged as low risk.”  

ETH Zurich, a public university in Zurich, Switzerland, offers a variety of courses in AI ethics, policy, and governance. Course titles include Philosophy of Language and Computation, GenAI in Teaching and Learning, Health Data and AI: Responsible Innovations, Ethics and Regulatory Strategies, and Language Models for Law and Social Science. 

In Italy, the Universities of Milan, Milano-Bicocca and Pavia offer a Joint Master’s Degree in Human-Centered Artificial Intelligence. The degree includes three programs – Hybrid AI, Neuro-AI, and AI and Law – with compulsory courses including AI and Human Decision-making, and Brain and Cognition. Optional studies for the two-year degree include Knowledge Representation and Reasoning, Methodology of Data-Driven Reasoning, Natural Language Processing, and AI Applied to Neurological Sciences and Brain-Computer Interfaces. 

AI ethics courses aren’t limited to universities. The subject regularly appears in the curriculum for primary and secondary school students, and courses are also available for the general public.

The Computer Science Teachers Association (CSTA) promotes ethical AI training for primary and secondary school students through role-playing and discussion. In one example of a six-session lesson plan, students begin by watching the 2021 animated science fiction film Ron’s Gone Wrong, in which children use robots to make friends by algorithms that parallel real-life social media. In the film, the malfunction of one robot teaches the film’s characters about the true nature of human friendship.  

In the first of the six lessons, each student selects one of the film’s characters, which are based on four archetypes (the Outsider, the Influencer, the Bully, and the Traditionalist). Watching the film from the perspective of their chosen character, the students witness the consequences of AI on social dynamics while documenting moments that parallel real life. Working individually and in groups through the lessons, the students build, present, and compare arguments and viewpoints from the perspective of their chosen character. 

One computer science teacher who implemented the “Ron’s Gone Wrong” lesson plan in his 6th-grade class observes that “Students can discuss controversial concepts or question assumptions, while holding onto a little distance by role-playing as their character. A student might hesitate to say, ‘AI algorithms favor wealthy people’ in their own voice, but as Barney [the Outsider] or Rich [the Bully], they can explore that awkward truth with greater freedom.”

 

 

Courses in AI ethics aren’t limited to students enrolled in schools and universities. Several courses are freely available to the general public through schools like the University of Helsinki and platforms like Coursera, while other offerings are targeted to specific professions. For example, in April the United Nations Educational, Scientific and Cultural Organization (UNESCO) and the University of Oxford announced the launch of a free Massive Open Online Course (MOOC) entitled AI, Justice and Rule of Law, designed to help justice professionals navigate the ethical, legal and human rights challenges of AI. 

What makes AI “ethical”?

Ethical interpretations differ widely around the globe, making universal agreement on what’s ethical and what’s not elusive and complicating the design of AI ethics training. While specific ethical practices vary by region and culture, in 2021 the 193 nation members of UNESCO reached a general agreement on ethical boundaries. 

UNESCO members have agreed on four core values for ethical AI: 

• To respect and promote human dignity and human rights 

• To foster just, peaceful, and interconnected societies 

• To ensure diversity and inclusiveness 

• To support flourishing of environment and ecosystems. 

These core values are backed by ten principles that include inarguably ethical terms like “do no harm,” “safety,” “transparency,” “privacy,” “governance,” “responsibility,” “sustainability,” and “fairness.” 

 

 

One of the “godfathers of AI,” Yoshua Bengio, speaks with UNESCO on the question of whether humans will be able to control AI.

Bengio advocates for AI to be “managed like a global public good.” Read The Quantum Record’s September 2025 feature on Bengio, his non-profit LawZero, and proposal for “Scientist AI”: ‘Godfather of AI’ Proposes New Type of Neural Network to Guard Against LLM Deceit and Self-Preservation 

In 2022, UNESCO applied these core values and principles to its Recommendations on the Ethics of Artificial Intelligence (available in pdf in various languages). The practical limitations of the recommendations were acknowledged in a caveat: “This Recommendation does not have the ambition to provide one single definition of AI, since such a definition would need to change over time, in accordance with technological developments. Rather, its ambition is to address those features of AI systems that are of central ethical relevance. Therefore, this Recommendation approaches AI systems as systems which have the capacity to process data and information in a way that resembles intelligent behaviour, and typically includes aspects of reasoning, learning, perception, prediction, planning or control.” 

Although the definition of “intelligent behaviour” is likely to be forever contentious, three elements underpin UNESCO’s approach to the relationship of ‘artificial’ and ‘real’ intelligence. Recognizing that the term “AI” is now widely used without general agreement on what makes it different from any other computer process, UNESCO’s elements begin with a definition that encapsulates the autonomous nature of the general term: “AI systems are information-processing technologies that integrate models and algorithms that produce a capacity to learn and to perform cognitive tasks leading to outcomes such as prediction and decision-making in material and virtual environments.”  

This distinction of “Artificial” from “Real” intelligence, often assumed but seldom discussed, is the key that leads to UNESCO’s four core values. Each of the core values speaks to human flourishing as the ultimate goal, which often stands in opposition to the immediate economic incentives of commercial exploitation. Logically, “artificial” intelligence exists because of its contrast to and derivation by “real” human intelligence, making the objective of human flourishing of prime importance.

 

Transmitting intelligence is the responsibility of parents, teachers, and social leaders. Training teachers of all levels from elementary to university in AI ethics is a major challenge for institutions.

As increasing numbers of students are exposed to AI ethics courses, it is crucially important that their teachers also receive training. Teacher training in AI ethics is advancing in a few different ways. 

For example, an article presented at the Thirty-Seventh AAAI Conference on Artificial Intelligence in 2023, entitled Literacy and STEM Teachers Adapt AI Ethics Curriculum (pdf available), discusses the success of a pilot program that helps teachers to learn and incorporate AI ethics in a wide range of subjects. 

The authors, led by Benjamin Walsh from the Department of Computer Science at the University of Colorado Boulder, coordinated a team of educators and researchers in STEM and literacy studies who designed a project-based AI ethics curriculum for middle and high school students. Curriculum modules included topics such as algorithmic bias, conflicting goals in privacy and security, and the ethics of AI-created art and self-driving cars. 

The modules were designed for teachers to deliver the material in a variety of different ways, including the use of original short stories, non-fiction texts, and hands-on AI experience such as the design and sharing of digital media, to help students develop a holistic understanding of AI ethics. 

The authors studied the ways that a 7th grade computer science teacher, two 9th grade English Language Arts teachers, and a 9th grade cybersecurity teacher customized and applied the AI ethics curriculum for the particular circumstances of their student groups from different cultures in a variety of urban, rural, and suburban communities. The study’s results suggest that even without previous formal training in AI ethics, teachers were able to adapt the modules and their flexible delivery options to respond to the needs of different student groups while adhering to school guidelines. 

The study concludes that “While it would be possible to offer teachers a more scripted, ‘teacher proof’ AI ethics curriculum, we believe this is an insufficient solution. There is no one-size-fits-all approach for teaching AI ethics. On the one hand, teachers will encounter students with a wide range of CS [computer science] backgrounds. On the other, they teach in different communities and settings where views on contentious social issues can differ dramatically.” 

As AI’s capabilities and reach expand, what is on the horizon for AI ethics training and implementation?

The question is difficult to answer, particularly as regulators around the globe grapple with uncertainty, entrenched commercial interests, and the speed of AI development that outpaces regulatory processes.

 

 

With its AI Act implemented in 2024, the European Union and its 27 member nations are at the forefront of a global movement to defend public interests in the battle for ethical AI development. The EU’s AI Act approaches ethics by degrees of risk, categorizing AI applications as either minimal risk, limited risk, high risk, or unacceptable risk. Applications in the latter category are prohibited, including social scoring (which has been implemented in China) and creation of facial recognition databases through untargeted scraping of the internet and CCTV footage. 

The AI Act is one of many subjects discussed in AI ethics courses and programs around the globe. Although the courses and programs are relatively new, the rapid increase in student interest and teaching resources holds promise for widespread practical implementation of ethical principles in the evolving design of AI applications. 

The answer to the question of what’s next on the horizon for AI ethics may lie in the hands of the commercial interests developing the applications. Will the proliferation of AI ethics training demonstrate the commercial benefits of implementing sustainable AI practices? Maybe only time will tell. 

---

Craving more information? Check out these recommended TQR articles:

• Fraudulent Data Influencing Decision-Making: AI, Fake Respondents, and the Future of Public Opinion Research 
• Research Highlights Benefits of AI Literacy and Questioning Outputs to Combat Cognitive Offloading 
• The Algorithmic Governance Challenge: Inside the Battle Over Social Media Algorithms