Insights Gleaned from FPF's "AI Systems Implementation" Seminar
In a significant step towards navigating the rapidly evolving landscape of Artificial Intelligence (AI), the Future of Privacy Forum (FPF) hosted a thought-provoking workshop titled "Deploying AI Systems" as part of the Privacy + Security Academy's Spring Academy. The event, which took place at The George Washington University in Washington, DC on May 7, 2025, attracted students, privacy lawyers, and representatives from various firms, companies, data protection authorities, and regulatory agencies worldwide.
The workshop, facilitated by Amber Ezzell, Policy Counsel for Artificial Intelligence at FPF, along with Anne Bradley (Luminos.AI), Brenda Leong (ZwillGen), Bret Cohen (Hogan Lovells), and Daniel Berrick (FPF), aimed to explore emerging U.S. and global legal requirements for AI deployers. The key takeaways from the workshop emphasised the importance of balancing innovation, privacy protection, governance, and competition as AI systems are deployed in diverse contexts, with a strong focus on risk management and regulatory challenges.
One of the core themes of the workshop was the immense opportunities from agentic AI, which offers significant advances and potential benefits. However, these come with complex privacy, governance, and policy challenges that must be thoughtfully addressed during development. Risk management was identified as a top priority for organisations developing agentic AI, with panelists such as Snively emphasising that the potential rewards of deploying AI systems are likely to outweigh the risks if managed properly.
The critical intersection between privacy and competition in the AI era was another topic of discussion. Regulators were urged to empower users to protect their privacy while ensuring fair competition in the market. The workshop highlighted that antitrust policies favoring openness can create privacy and security risks, so regulators must treat privacy enforcement as an element of market competition.
The workshop also addressed the issue of some privacy laws unintentionally hindering competition, with AI becoming a frequently cited element in privacy debates—even when it might not directly apply. This calls for a nuanced understanding and application of privacy laws in the context of AI.
Participants agreed that deployers of AI tools should independently test tools using their own data for high-risk use cases, rather than relying on third-party representations. Agentic systems, while not currently a main focus of risk governance for most participants, were identified as a potential area that may soon pressure test or amplify governance questions relevant to more widely deployed forms of AI.
Organisations are investing more resources towards AI adoption and innovation, but agree on the importance of a risk-based approach to AI deployment for mitigating risk and regulatory pitfalls. The workshop concluded with a reminder that FPF will continue to explore a range of issues at the intersection of AI governance as organisations, policymakers, and regulators grapple with the challenges of AI development and deployment.
Regulators are showing growing interest in unfair and deceptive practices related to AI deployment, including misleading statements about capabilities, implementation, and data management practices. The workshop provided a valuable platform for discussing these issues and fostering collaboration to ensure the responsible and ethical deployment of AI systems.
- The Future of Privacy Forum (FPF) hosted a workshop titled "Deploying AI Systems" to discuss emerging legal requirements for AI deployers on a global scale.
- The workshop highlighted the importance of balancing innovation, privacy protection, governance, and competition while deploying AI systems in diverse contexts, with a strong focus on risk management and regulatory challenges.
- Participants agreed that deployers of AI tools should independently test tools using their own data for high-risk use cases, rather than relying on third-party representations, and that agentic systems could soon pressure test governance questions relevant to more widely deployed forms of AI.
- Regulators are showing growing interest in unfair and deceptive practices related to AI deployment, including misleading statements about capabilities, implementation, and data management practices, and the workshop provided a valuable platform for discussing these issues and fostering collaboration to ensure the responsible and ethical deployment of AI systems.
