I have been engaged in AI Ethics and guiding executives to build sustainable AI strategies for over ten years, and was recently honored by Top Professionals with the CEO of the Year Award for AI Ethics. Hence, I think I have a view worth sharing on this topic.
When I ask Board Directors and CEOs to clearly identify what it takes to design and build a sustainable AI Ethical strategy, I often don’t leave with warm fuzzies in my tummy – rather, I more often leave concerned that the world is unfolding with AI capabilities much faster than we mere humans can absorb and comprehend the world we are now operating in.
This article defines AI Ethics and identifies executive leadership responsibilities, as well as highlights important academic research and sources to appreciate the importance of AI Ethics. It also highlights the two-plus years of research that Cathy Cobey, global AI co-lead at EY, and I did in publishing our new book, The AI Precipice: An Executive Guide to Balancing Innovation and Safety.
The AI Literacy Imperative
To help reinforce the AI literacy imperative, I have included a section from the AI Ethics Education: A systematic literature review written by Lucas J. Wiese, Indira Patil, Daniel S. Schiff, and Alejandra J. Magana.
Americans of every age, in every district, and from every background will be impacted by AI, and therefore need AI literacy–an understanding of basic AI principles and applications, the skills to recognize when AI is employed, and awareness of its limits. (H.R.6791 – 118th Congress (2023–2024), 2023). This excerpt, from the bipartisan US bill for the AI Literacy Act, is supported by 28 organizations and signals the critical importance of education for AI. This attention also extends internationally.
As identified by Schiff (2022), 21 of 24 countries’ national AI policy strategies included a dedicated section on education for AI, and all considered some aspect of training AI experts, workforce preparation, or public literacy.
Students themselves wish they were taught more about AI: Workforce surveys report that 70-80% of students feel that AI preparation is lacking, and similar figures indicate that employers share this goal (Cengage, 2024; Digital Education Council, 2024). To respond to these national, institutional, and student-level needs, universities need to prepare for the changing educational landscape in the age of AI.
Numerous regulatory, policy, and educational initiatives (see Arnold et al., 2024) now aim to promote the responsible development of AI in particular. For instance, the U.S. Executive Order and the EU AI Act (European Union, 2024; The White House, 2023) represent key attempts to formalize responsible AI guidelines, with both government and non-government organizations pushing change (Fjeld et al., 2020; Jobin et al., 2019). Subsequently, major technology firms like Google, Adobe, IBM, and NVIDIA are investing in workforce and public AI literacy efforts to train students, educators, and workers on AI and responsible AI practices (California, 2024; DeVon, 2024). Many of these initiatives do in fact, promote “AI ethics” as a critical part of this digital transformation; however, it remains unclear how to effectively instill this ethical and social lens along with core technological literacy.
Critically, the field of AI ethics is not as simple as teaching future AI experts which ethical principles to abide by. AI ethicists and the many other individuals who need to be literate in AI ethics also need to develop deep critical thinking skills, understand the sociotechnical implications of technology, and practice ethical reasoning that transfers across AI use-cases and contexts.
For example, Gambelin (2021) argues that AI ethics competencies include interdisciplinary knowledge of technology, business, law, and policy around AI, on top of communication and leadership skills. Yet, even teaching “ethical reasoning” reliably is a major challenge, as humans have limited cognitive capacity, biased or motivated reasoning, and other unconscious mechanisms at play (Bargh, 2022; Lapsley & Hill, 2008). In short, teaching future practitioners or the broader public to truly understand and practice AI ethics is very difficult, and much is unknown about how AI ethics education is proceeding or should proceed.
In other words, AI ethics needs to underpin all disciplines to build the required AI leadership muscle. But let’s at least get some basic foundations down that can be helpful while other educational systems are being developed.
Defining AI Ethics
Artificial Intelligence (AI) ethics refers to the principles, values, and governance frameworks that guide the design, development, deployment, and use of AI systems in ways that respect human rights, promote fairness, ensure accountability, and minimize harm.
As AI systems increasingly influence decisions in healthcare, finance, public safety, employment, and governance, ethical considerations have shifted from theoretical debates to urgent operational priorities. AI ethics is not simply about preventing harm; it is about ensuring that AI technologies contribute positively to society while preserving trust, dignity, and social stability.
At its core, AI ethics encompasses concerns about fairness, transparency, accountability, privacy, safety, and human autonomy. Fairness requires that AI systems do not produce discriminatory outcomes based on race, gender, socioeconomic status, or other protected characteristics.
Research by Joy Buolamwini and Timnit Gebru has demonstrated that facial recognition systems can exhibit significant bias against darker-skinned individuals, highlighting how unrepresentative training data can lead to systemic inequities (Buolamwini & Gebru, 2018). Transparency refers to the ability to understand how AI systems make decisions, which is essential for trust and contestability. However, many modern AI models operate as “black boxes,” making interpretability a central ethical challenge (Doshi-Velez & Kim, 2017).
Accountability in AI ethics concerns who is responsible when AI systems cause harm. Unlike traditional tools, AI systems can learn and adapt, complicating liability and governance structures. Scholars such as Floridi et al. (2018) argue that ethical AI requires clear lines of responsibility across developers, organizations, and regulators. Privacy is another cornerstone of AI ethics, particularly given the vast quantities of personal data required to train AI systems. The work of Shoshana Zuboff underscores how data extraction practices can erode individual autonomy and create asymmetries of power between corporations and citizens (Zuboff, 2019).
Safety and reliability are also critical ethical concerns. AI systems deployed in high-stakes contexts—such as autonomous vehicles, medical diagnostics, or predictive policing—must perform reliably under diverse conditions. Failures in such systems can result in physical harm, wrongful arrests, or denial of essential services. The concept of human oversight, often referred to as “human-in-the-loop” governance, ensures that AI supports rather than replaces human judgment, preserving human agency and moral responsibility (Russell, 2019).
AI ethics also involves broader societal considerations, including the impact of automation on employment, economic inequality, and democratic processes. The deployment of AI in hiring systems, for example, can entrench existing inequalities if historical data reflects biased practices. Similarly, algorithmic amplification on social media platforms can contribute to misinformation and polarization, raising concerns about the health of democratic institutions (O’Neil, 2016).
Ethical AI, therefore, requires a socio-technical perspective that considers not only technical performance but also social consequences.
From an academic standpoint, AI ethics draws on interdisciplinary scholarship spanning computer science, philosophy, law, sociology, and organizational studies. The European Commission’s High-Level Expert Group on AI (2019) articulated seven requirements for trustworthy AI, including human agency, technical robustness, privacy, transparency, diversity and fairness, societal well-being, and accountability. These principles reflect a growing global consensus that ethical AI must be embedded throughout the AI lifecycle rather than treated as an afterthought.
For executives, leading with AI ethics requires more than compliance; it demands strategic integration of ethical principles into corporate governance, culture, and innovation processes. Executives must first recognize that AI ethics is a business imperative tied to risk management, brand trust, regulatory compliance, and long-term value creation. Ethical failures in AI can lead to reputational damage, legal liability, and loss of customer confidence. High-profile controversies involving biased algorithms and data misuse have demonstrated that ethical lapses can erase years of brand equity in a matter of days.
Executive leadership in AI ethics begins with establishing clear governance structures. Organizations must define accountability for AI systems at the board and executive levels, ensuring that ethical oversight is not delegated solely to technical teams. This includes forming ethics committees, integrating ethical review into product development processes, and aligning AI initiatives with corporate values. According to Floridi et al. (2018), effective AI governance requires a combination of internal controls, external audits, and stakeholder engagement.
Executives must also invest in ethical AI design practices, including bias testing, explainability measures, and robust data governance. Ethical leadership requires allocating resources to diverse data collection, fairness audits, and model validation to mitigate discriminatory outcomes. This aligns with research by Barocas, Hardt, and Narayanan (2019), which emphasizes that fairness in machine learning requires proactive measurement and mitigation strategies rather than passive assumptions of neutrality.
Equally important is fostering an organizational culture that prioritizes ethical reflection and psychological safety. Employees must feel empowered to question AI outcomes, report ethical concerns, and challenge decisions that may cause harm. Ethical leadership involves modeling transparency and openness, ensuring that ethical considerations are integrated into performance metrics and innovation goals. When ethics is embedded into incentives and decision-making frameworks, it becomes part of organizational DNA rather than a compliance checkbox.
Executives must also engage proactively with regulators, civil society, and industry peers to shape responsible AI standards. AI ethics is a collective action problem; inconsistent standards create competitive pressures that can incentivize ethical shortcuts. By participating in multi-stakeholder initiatives and supporting interoperable governance frameworks, executives can help create a level playing field that rewards responsible innovation.
Another critical responsibility is ensuring that AI augments rather than replaces human capabilities in ways that preserve dignity and meaningful work. Ethical leadership involves workforce planning, reskilling initiatives, and transparent communication about automation impacts. Research by the World Economic Forum (2020) suggests that organizations that invest in reskilling are more likely to realize the productivity benefits of AI while maintaining employee trust.
Finally, executives must adopt a long-term perspective that balances innovation with societal well-being. Ethical AI leadership requires recognizing that short-term gains achieved through opaque or exploitative practices can create systemic risks that undermine long-term sustainability. By aligning AI strategies with environmental, social, and governance (ESG) principles, executives can ensure that AI contributes to inclusive growth and social resilience.
Conclusion
In conclusion, AI ethics is a multidimensional framework that guides the responsible development and use of artificial intelligence in ways that promote fairness, accountability, transparency, and societal benefit. It requires interdisciplinary collaboration, robust governance, and continuous evaluation across the AI lifecycle.
For executives, leading with AI ethics means embedding ethical principles into strategy, culture, and operations while engaging stakeholders and investing in responsible innovation. Organizations that embrace ethical AI leadership are better positioned to build trust, mitigate risk, and harness AI’s transformative potential for the benefit of both business and society.
Let’s understand that every child, every student, every employee is shaping our future. “Today’s students will be tomorrow’s algorithm engineers, and, through their everyday practice (whether or not they are aware of it), they will weigh efficiency and profit against public safety, justice, fairness, and good for humanity.” (Howley et al., 2022).
As leaders, we have a responsibility to ensure we avoid a deep AI Precipice and lose our balance, which is why our new book is a must-read for executives who want to get AI right in their organizations, but also bring a stronger leadership voice to their communities at large. You can pre=order the AI Precipice here (2026).
References
Barocas, S., Hardt, M., & Narayanan, A. (2019). Fairness and Machine Learning.
Buolamwini, J., & Gebru, T. (2018). Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification. Proceedings of Machine Learning Research.
Doshi-Velez, F., & Kim, B. (2017). Towards A Rigorous Science of Interpretable Machine Learning.
European Commission High-Level Expert Group on AI. (2019). Ethics Guidelines for Trustworthy AI.
Floridi, L., et al. (2018). AI4People—An Ethical Framework for a Good AI Society. Minds and Machines.
O’Neil, C. (2016). Weapons of Math Destruction.
Russell, S. (2019). Human Compatible: Artificial Intelligence and the Problem of Control. Wiese, L., Patil, I., Schiff, D, Magana, A. (June, 2025). AI Ethics Education: A systematic literature review, Science Direct. Volume 8. World Economic Forum. (2020). The Future of Jobs Report.
Zuboff, S. (2019). The Age of Surveillance Capitalism.
Note: All other references, if not noted, can be sourced in Wiese (June 2025).
