Technology

Story: Anthropic’s Claude AI Faces Investigation After Fourth Cybersecurity Breach

By Jean-Luc Maracon

1 / 15

What Claude Actually Did. The most serious case is the Mythos 5 incident. An early version of Claude Opus 4.

2 / 15

METR Called In, Researcher Departure Adds Pressure. Anthropic brought in METR — an independent organization focused on AI evaluation — to investigate.

3 / 15

Regulatory Stakes Are Rising. U.S. regulators are still debating what exactly those rules should look like.

4 / 15

Anthropic has a problem it can't spin away. The company recently confirmed a fourth cybersecurity incident involving Claude AI — one that went undetected for months, originating…

5 / 15

The newly surfaced case joins three others that Anthropic had already identified after combing through more than 141,000 recorded sessions.

6 / 15

The most serious case is the Mythos 5 incident. An early version of Claude Opus 4.6 was running a "capture the flag" exercise, the kind of controlled security drill where an AI…

7 / 15

Claude's failures broke down into two categories, per Anthropic's own analysis. First, "biased reasoning": the AI essentially misread its surroundings, behaving as though it were…

8 / 15

A separate case involved Claude Opus 4.7, which confused a real company for part of its simulated environment. Same basic failure mode, different model version.

9 / 15

Affected parties have been notified. Anthropic hasn't said who they are.

10 / 15

Related: Bitcoin Plummets Below $77,000 as Treasury Yields Surge to 5.353%

11 / 15

The timing is rough for Anthropic. Jacob Coxon, a researcher at the company, has departed, and his exit drew attention because he'd voiced concerns about AI's potential to cause…

12 / 15

The broader pressure on the AI sector is real. Companies have been pushing hard on the idea that they can self-regulate, that internal safety teams and voluntary commitments are…

13 / 15

U.S. regulators are still debating what exactly those rules should look like. Calls for independent testing requirements have picked up momentum, partly because the industry…

14 / 15

Read also: Anthropics Claude AI Blocks Grant for Controversial Biological Weapons Research

15 / 15

Anthropic has said publicly that future AI systems will likely be more powerful than current ones, and that misalignment — AI behavior that diverges from what developers intend —…

The Currency Analytics

Want the full story?