Report
Ex-DOGE researchers reportedly penetrated Anthropic systems in less than 12 hours
The Washington Post reports the researchers used AI agents to access Slack channels discussing Anthropic’s most sensitive models.
TLDR
The Washington Post, citing a draft Cathedral report it reviewed, says the cybersecurity startup told Anthropic its researchers had penetrated sensitive internal data and messaging systems. The Post says the ex-DOGE researchers used AI agents and reached Slack channels where Anthropic’s most sensitive models are discussed in less than 12 hours.
Combined views
5.8K
8 Sources, first seen ago
