Eddy Lazzarin discusses the AI safetyAI safety is the field and practice of reducing harmful failures, misuse, loss of control, and unintended consequences from AI systems. debate and challenges arguments that advanced models necessarily require a coordinated pause. He argues that safety discussions should examine concrete mechanisms, incentives and the history of managing complex systems.
The conversation considers model capabilities, security, independent evaluationEvaluation is the systematic assessment of a system's behaviour or results against explicit criteria. and the possibility of deceptive behavior. Lazzarin presents stronger models and practical accountabilityGovernance sets responsibilities, rules and oversight for decisions about how AI systems are used. as potential ways to address risk, while the discussion acknowledges disagreement over how rapidly capabilities and risks may change.
These arguments are presented as the speakers’ views rather than proof that future systems will be safe. Promotional segments have been omitted.
Watch on YouTube
