Key AI Safety Sessions Highlighted at TechCrunch Disrupt 2026
TechCrunch Disrupt 2026 has announced five critical sessions focusing on enterprise AI deployment, agent security, cloud infrastructure, and robotic systems.
With AI technologies moving past the testing phase and integrating into businesses, tools, and robots, security and reliability have become critically important. Five special sessions at TechCrunch Disrupt 2026 will enable founders to tackle challenges in this domain.
Enterprise Deployment and the Anthropic Experience
While some businesses derive tangible value from AI, others remain stuck in the experimentation phase for months. Success and failure factors behind this process are being examined.
Cat de Jong, Head of Applied AI at Anthropic, will share with attendees what happens when companies transition AI from experimental phases to production processes.
AI Agent Security
AI agents possessing the ability to take action independently bring entirely new security problems with them.
How to ensure agent security at the infrastructure level when application-level permissions fall short is under discussion.
Cloud Infrastructure and Secure Enterprise Systems
Getting an AI product accepted at the enterprise level requires much more than an impressive demo.
Security, governance, and observability sit at the center of processes where companies consider using AI in critical systems.
Zero-Tolerance-for-Error AI in Physical Systems
AI systems operating in the physical world directly impact vehicles, aircraft, and industrial processes.
From the perspective of defense and autonomous systems, how to determine whether an autonomous system is ready for deployment is being addressed.
The Data Problem of the Robotics World
Robots struggle to access large training data pools that have accelerated the development of language models and autonomous vehicles.
Data pipelines, simulation environments, and how foundation models can bridge this gap are being evaluated by experts.