2. Security

Google security team members Priya Pandey, Aron Eidelman, and Leonid Yankulin explain why probabilistic AI is breaking traditional DevSecOps, and how runtime security and safer observability must evolve to keep sensitive data out of your logs.

New platform focuses on runtime enforcement, auditability, and risk scoring for AI agents operating in regulated enterprise environments.

The Pentagon’s blacklisting of Anthropic over AI weapons and surveillance restrictions exposes a new class of governance risks for enterprise AI — and reveals how differently America’s leading AI companies view their obligations to the state.

Platform updates target automation, continuous validation, and lifecycle protection for AI-driven systems.

Security models are no longer enough as multi-modal attacks overwhelm traditional controls, forcing a rethink of enterprise trust systems.

The Promptfoo deal underscores the importance of model evaluation, red-teaming, and reliability in scaling enterprise AI deployments.