Shield AI, Waabi, and General Motors on building AI when failure is not an option at TechCrunch Disrupt 2026

The companies involved in this discussion are building AI for situations where errors can have catastrophic consequences, signaling that rigorous validation and safety culture are becoming paramount for all AI deployments.

Leaders from Shield AI, Waabi, and General Motors are scheduled to discuss the challenges of developing artificial intelligence systems where failure is not an option at TechCrunch Disrupt 2026. The panel, titled "Building AI Systems When Failure Is Not an Option," will focus on deploying AI in high-stakes environments, including defense, autonomous vehicles, and industrial robotics. Speakers will cover crucial aspects such as establishing a safety culture, rigorous testing and validation processes, navigating regulatory frameworks, and building public trust. Shield AI's Chief Technology Officer, Nathan Michael, brings experience from deploying AI for the U.S. Air Force. Waabi founder and CEO, Raquel Urtasun, is known for her work in autonomous driving technology and her company's sophisticated simulation tools. General Motors' Director of Robotics Strategy, Mikell Taylor, offers insights from developing practical and reliable robots for human collaboration. The discussion aims to address how to determine when AI systems are truly ready for real-world operation, moving beyond the lab to critical applications.

This panel highlights a critical inflection point for AI development, moving from theoretical advancements to practical, safety-critical applications. The focus on "when failure is not an option" underscores the increasing demand for verifiable AI, particularly as autonomous systems enter regulated industries and public spaces. Companies are investing heavily in simulation, validation, and regulatory compliance, indicating that market entry for robust AI will require demonstrating a high degree of safety and reliability.

The dialogue between defense, automotive, and industrial robotics leaders suggests a convergence of best practices for AI safety across diverse sectors. It implies that companies in less critical domains may also need to adopt similar stringent development and testing methodologies to gain user and regulatory trust. The emphasis on user experience and human-robot interaction, particularly from GM's perspective, indicates that the deployment of AI will increasingly depend on its ability to integrate seamlessly and safely with human environments.