AI Verification Mechanisms and d/acc Happy Hour: GPAI + Halcyon + SeIdon IASEAI Side Event

​Side event to the IASEAI conference, hosted by the General Purpose AI Policy Lab, Halcyon Futures, and Seldon Lab.

​As AI systems become more capable, how do we ensure defenses keep pace with advancing capabilities? This evening focuses on Verification Mechanisms & other defensive acceleration (d/acc) approaches that can provide guarantees about AI systems' training, capabilities, and safety properties.

​This informal gathering brings together researchers, policymakers, and practitioners working on technical and governance challenges: What can we realistically verify or control today? What would more robust mechanisms require? And how should these strategies shape AI policy?

​Cocktails at Gatsby, 7th arrondissement. We look forward to seeing you there.

​

​—

​Halcyon Futures is a San Francisco-based AI resilience fund. We make career transition grants, nonprofit incubation grants, and early-stage venture investments in exceptional entrepreneurs focusing their life's work on mitigating societal risks from advanced AI. We're early backers of orgs and companies like Transluce, AI Verification and Evaluation Research Institute (AVERI), Goodfire, Artificial Intelligence Underwriting Company, and Asymmetric Security.

​The General-Purpose AI Policy Lab is a non-profit research organization working alongside French (and European) institutions to address the security and international coordination requirements posed by the development of general-purpose AI (GPAI).

​Seldon Lab brings together the world's most capable founders while conducting our own cutting-edge research and directly developing the existential security technologies that will define how humanity related with general machine intelligence. We believe AGI security will become the second largest industry in history.