Security firm Codewall turned an offensive AI agent loose on McKinsey's internal AI platform Lilli, a system used by over 43,000 employees for strategy work, client research, and document analysis. No credentials, no insider knowledge, no human assistance. Within two hours, the agent had full read and write access to the production database.

Google's first native multimodal embedding model brings text, images, video, audio, and documents into one vector space, cutting out the need for separate models in AI pipelines.

The AI tool has become the country's latest tech obsession. For savvy early adopters, that's a business opportunity.

OpenAI has released IH-Challenge, a training dataset designed to teach AI models to reliably prioritize trusted instructions over untrusted ones. Early results show significant improvements in both security and prompt injection defense.

A broad coalition is backing Anthropic in its legal fight with the US Department of Defense: Microsoft, dozens of OpenAI and Google employees, former military leaders, and civil rights organizations have all filed amicus curiae briefs supporting the AI company.