It took just 12 hours for an AI start-up founded by ex-DOGE staffers to break into Anthropic

Ex-DOGE staffers break into Anthropic. (fixed it)

In less than 12 hrs, cyber researchers were able to penetrate Anthropic’s systems using AI agents—including Slack channels where the most sensitive models are discussed. Just weeks after whistleblower Joseph Coxon charged the company w/launching “a hubristic gamble” from a “private company’s Slack.”

Exclusive: Cathedral, a months-old cybersecurity start-up, recently told Anthropic that its researchers had penetrated sensitive internal data and messaging systems at the AI company, according to a draft report reviewed by The Post.

Terrifying www.washingtonpost.com/technology/2...

People who read my substack post in June (link in next post) about how Anthropic & Open AI's frontier models would be easy to steal & were probably already stolen by multiple state actors will not be surprised one bit about this discovery www.washingtonpost.com/technology/2...

Cathedral, a months-old cybersecurity startup founded by ex-DOGE staffers, says it breached Anthropic's internal systems in under 12 hours using AI agents (Washington Post) Main Link | Techmeme Permalink

Cathedral, a months-old cybersecurity start-up, recently told Claude chatbot maker Anthropic that its researchers had penetrated sensitive internal data and messaging systems at the artificial intelligence company. www.washingtonpost.com/technology/2...

Has 2026 become the 'Year of Autocannibalism'? It took just 12 hours for an AI start-up to break into Anthropic www.washingtonpost.com/technology/2...

seems bad

www.washingtonpost.com/technology/2... Did anyone really believe these former DOGE guys would fade into anonimity? When Dems win both houses, they should start by investigating, convicting & imprisoning these guys. It will be easier to do so than to do so first to their fearless treasonous leader

The young hackers that Trump allowed Musk to unleash on our personal & government databases have no allegiance to America. They, like Trump & Musk, are in it for themselves. Take precautions: keep records of your financial accounts, change your passwords, VOTE!

So that's what DOGE was doing? www.washingtonpost.com/technology/2...

I'd like to meet the brilliant financial or legal mind at Anthropic who thought this would make a great first amendment to their registration statement. I would have put it in the first draft but hey what do I know? www.washingtonpost.com/technology/2...

It took just 12 hours for an AI start-up founded by ex-DOGE staffers to break into Anthropic (Washington Post) Main Link | memeorandum Permalink

Kyberbezpečnostní startup Cathedral využívající AI tvrdí, že pronikl do datových a komunikačních systémů společnosti Anthropic www.washingtonpost.c...

Siber güvenlik girişimi Cathedral'ın araştırmacıları, Anthropic tarafından kullanılan üçüncü taraf kurumsal yazılımlardaki güvenlik açıklarından yararlanarak bazı şirket içi sistemlere erişti. ... #Anthropic #SiberGuvenlik #YapayZeka #YZ #AI 🔗 www.washingtonpost.com/technology/2...

“The start-up firm Cathedral’s ability to penetrate sensitive programs at one of the nation's leading AI companies comes amid mounting concerns about security in the global AI race.” www.washingtonpost.com/technology/2...

www.washingtonpost.com/technology/2...

www.washingtonpost.com/technology/2...

This was a security test. Anthropic says it blocked access and found no evidence of other exploitation. Claude chats and model weights weren't reached. www.washingtonpost.com/technology/2...

How an AI start-up founded by ex-DOGE staffers broke into Anthropic The start-up firm Cathedral’s ability to penetrate sensitive programs at one of the nation's leading AI companies comes amid mounting concerns about security in the global AI race.

Researchers at cybersecurity start-up Cathedral got inside Anthropic - The Washington Post

It sounds like Anthropic has shitty IT security & doesn't patch their business software, or left openings to the flawed applications, which led to a daisy chain attack/elevation. That's basic IT stuff, they failed. That's how their models leak, world ends. 🤦 www.washingtonpost.com/technology/2...

It took just 12 hours for an AI start-up founded by ex-DOGE staffers to break into Anthropic www.washingtonpost.com/technology/2...