Five AI safety sessions every founder should have on their TechCrunch Disrupt 2026 agenda
techcrunch ·
This article highlights five essential AI safety sessions for founders attending TechCrunch Disrupt 2026, emphasizing that trust and security are now critical components of product deployment. The first session features Anthropic’s Cat de Jong discussing the challenges enterprises face when moving from AI pilots to actual production deployments. Another key talk addresses agent security infrastructure with speakers from Okta and NanoCo, focusing on permission models for autonomous agents. AWS and cybersecurity experts will also examine the complex cloud security and governance requirements necessary for enterprise AI adoption. For hard-tech founders, a panel on physical AI explores how to ensure safety in robotics and autonomous systems where failure has real-world consequences. Finally, NVIDIA’s Les Karpas discusses the data gaps hindering robotics and how simulation and foundation models might help achieve a 'ChatGPT moment' for robots. These sessions aim to help builders understand how to earn customer trust amidst increasing regulatory and technical scrutiny.
techcrunch