Like Day and Night: Transformations in AI Safety Regulations from 2024 to 2025

Rebecca Jacobs, Associate General Counsel, Anthropic
Ben Rossen, CIPP/US, Global Head of AI Policy Legal, OpenAI
Jonathan Tam, CIPP/C, CIPP/US, Tech and AI Partner, Baker McKenzie

Artificial intelligence development continues at a breakneck pace, yet global regulatory approaches to AI safety risks — ranging from data breaches to mass surveillance — have become increasingly fragmented. The U.S. federal government has rescinded a prior executive order on AI safety, shifting toward a deregulatory stance aimed at maintaining a competitive edge in AI. While some U.S. states have pursued AI safety regulations, broad adoption throughout the country remains uncertain. Meanwhile, other jurisdictions, including the U.K., EU, and China, have either warned of regulation stifling innovation or advanced their own distinct frameworks. This panel will analyze regulatory changes over the past two years and their causes, and assess the roles of governments, businesses, consumer advocacy groups, and the public in ensuring AI remains safe and trustworthy into 2026 and beyond.

What you will learn:
• Understanding of how recent AI technological developments present significant safety challenges and opportunities.
• Strategies to comply with evolving AI safety regulations in the U.S., EU, U.K., China and other jurisdictions.
• Practical guidance on how to get involved in advocating for AI safety in different types of organizations.