The Road to Accountable AI

Holistic AI was one of the first companies built specifically to govern, audit, and red team AI systems. As co-founder and co-CEO Emre Kazim explains, its original thesis was that AI governance would mirror data governance: a compliance-driven regime. He now believes the better analogy is cybersecurity: a more technical, incident-driven discipline where best practices emerge from real-world events and propagate across industry, rather than descending from abstract regulatory frameworks. Kazim argues this shift has significant implications for who owns AI governance inside enterprises, what skills they need, and why documentation-and-reporting vendors are unlikely to capture the core of the market. Kazim also makes the case that human-in-the-loop oversight, long treated as the default answer to AI risk, has become untenable as systems grow more dynamic and agentic. He distinguishes between two enterprise adoption patterns: a democratic model in which every employee has a copilot, and a vanguard model in which a small number of mission-critical agentic systems drive most of the value and demand most of the governance attention. Finally, he argues that meaningful research capacity will be the price of entry for AI governance firms going forward. Dr. Emre Kazim is the co-founder and co-CEO of Holistic AI, an AI governance platform company spun out of University College London in 2020. He previously served as a Research Fellow in UCL's Department of Computer Science. Kazim has published more than 50 peer-reviewed articles on AI ethics and governance, serves as a member of the OECD's Network of Experts on AI, and is involved with the NIST AI Safety Institute. Transcript Towards Algorithm Auditing (Royal Society Open Science, 2024) What is AI Governance? (Holistic Blog, February 2026)

Former Congressman and Pentagon official Brad Carson discusses his organization, Americans for Responsible Innovation (ARI), which seeks to bridge the gap between immediate AI harms like and catastrophic safety risks, while bringing deep Capitol Hill expertise to the AI conversation . He argues that unlike previous innovations such as electricity or the automobile, AI has been deeply unpopular with the public from the start, creating a rare bipartisan alignment among those skeptical of its power and impacts. This creates openings for productive discussions about AI policy. Drawing on his military experience, Carson suggests that while AI will shorten the kill chain, it won't fundamentally change the human nature of warfare, and he warns against the US military's tendency to seek technical solutions to human problems . The conversation covers current policy debates, highlighting the necessity of regulating the design of models rather than just their deployment, and the importance of export controls to maintain the West's advantage in compute . Ultimately, Carson emphasizes that for AI to succeed politically, the "bounty" of this technology must be shared broadly to avoid tearing apart the social fabric Brad Carson is the founder and president of Americans for Responsible Innovation (ARI), an organization dedicated to lobbying for policy that ensures artificial intelligence benefits the public interest. A former Rhodes Scholar, Carson has had a diverse career in public service, having served as a U.S. Congressman from Oklahoma, the Undersecretary of the Army, and the acting Undersecretary of Defense for Personnel and Readiness . He also served as a university president and deployed to Iraq in 2008 . Transcript Former TU President Brad Carson Pushes for Strong AI Guardrails

Ravit Dotan argues that the primary barrier to accountable AI is not a lack of ethical clarity, but organizational roadblocks. While companies often understand what they should do, the real challenge is organizational dynamics that prevent execution—AI ethics has been shunted into separate teams lacking power and resources, with incentive structures that discourage engineers from raising concerns. Drawing on work with organizational psychologists, she emphasizes that frameworks prescribe what systems companies should have but ignore how to navigate organizational realities. The key insight: responsible AI can't be a separate compliance exercise but must be embedded organically into how people work. Ravit discusses a recent shift in her orientation from focusing solely on governance frameworks to teaching people how to use AI thoughtfully. She critiques "take-out mode" where users passively order finished outputs, which undermines skills and critical review. The solution isn't just better governance, but teaching workers how to incorporate responsible AI practices into their actual workflows. Dr. Ravit Dotan is the founder and CEO of TechBetter, an AI ethics consulting firm, and Director of the Collaborative AI Responsibility (CAIR) Lab at the University of Pittsburgh. She holds a Ph.D. in Philosophy from UC Berkeley and has been named one of the "100 Brilliant Women in AI Ethics" (2023), and was a finalist for "Responsible AI Leader of the Year" (2025). Since 2021, she has consulted with tech companies, investors, and local governments on responsible AI. Her recent work emphasizes teaching people to use AI thoughtfully while maintaining their agency and skills. Her work has been featured in The New York Times, CNBC, Financial Times, and TechCrunch. Transcript My New Path in AI Ethics (October 2025) The Values Encoded in Machine Learning Research (FAccT 2022 Distinguished Paper Award) - Responsible AI Maturity Framework

Kevin Werbach speaks with Caroline Louveaux, Chief Privacy, AI, and Data Responsibility Officer at Mastercard, about what it means to make trust mission critical in the age of artificial intelligence. Caroline shares how Mastercard built its AI governance program long before the current AI boom, grounding it in the company's Data and Technology Responsibility Principles". She explains how privacy-by-design practices evolved into a single global AI governance framework aligned with the EU AI Act, NIST AI Risk Management, and standards. The conversation explores how Mastercard balances innovation speed with risk management, automates low-risk assessments, and maintains executive oversight through its AI Governance Council. Caroline also discusses the company's work on agentic commerce, where autonomous AI agents can initiate payments, and why trust, certification, and transparency are essential for such systems to succeed. Caroline unpacks what it takes for a global organization to innovate responsibly — from cross-functional governance and "tone from the top," to partnerships like the Data & Trust Alliance and efforts to harmonize global standards. Caroline emphasizes that responsible AI is a shared responsibility and that companies that can "innovate fast, at scale, but also do so responsibly" will be the ones that thrive. Caroline Louveaux leads Mastercard's global privacy and data responsibility strategy. She has been instrumental in building Mastercard's AI governance framework and shaping global policy discussions on data and technology. She serves on the board of the International Association of Privacy Professionals (IAPP), the WEF Task Force on Data Intermediaries, the ENISA Working Group on AI Cybersecurity, and the IEEE AI Systems Risk and Impact Executive Committee, among other activities. Transcript How Mastercard Uses AI Strategically: A Case Study (Forbes 2024) Lessons From a Pioneer: Mastercard's Experience of AI Governance (IMD, 2023) As AI Agents Gain Autonomy, Trust Becomes the New Currency. Mastercard Wants to Power Both. (Business Insider, July 2025)