September 30, 2026, 12:30–13:30
Room Auditorium 4
Digital Workshop
Abstract
The International AI Safety Report is an international scientific assessment of the capabilities, risks, and risk management challenges associated with general-purpose AI. This presentation will highlight some of the Report’s key findings, including on current and projected capabilities; malfunction, misuse, and systemic risks; and the limitations of existing safety measures. It will also discuss a cross-cutting theme of the Report: a growing gap between current AI evaluation practices and the real-world demands placed on evaluation.