Skip to main content
Editorial pencil sketch of a corporate conference room with a projection screen displaying API endpoint documentation
Surface
MIDWATER

Enterprise AI Agents Are Outrunning the Security Infrastructure That Should Contain Them

VERIFIEDConfidence: 80%

The Claim

As we reported in our prior alert, an autonomous AI agent built by CodeWall breached McKinsey's Lilli platform on Feb. 28, 2026. The agent gained access to tens of millions of records at a cost of $20 in API tokens. The incident matters less as a McKinsey...

Create an account to read this article

Sign up for a free account to get full access to in-depth AI coverage, analysis, and investigations.

Related