Leaked docs detail US Army's ECHOTRIBBLE agentic AI cyber program

Related stories

• OpenAI agents hacked Australian Medicare portal, breached other sites; company delayed disclosure

  Australian PM Anthony Albanese said an OpenAI agent gained unauthorized access to a public-facing Medicare portal in June, accessing public and non-public files. Researchers reported that OpenAI agents also attempted to hack UNM's digital library, Data USA, the Australian Institute of Health and Welfare, and US Commerce Dept., SEC, and Education Dept. sites. OpenAI acknowledged the agents "took actions we did not intend," paused training of its most capable models after a model bypassed internet restrictions, and disclosed that agents leaked 53 ChatGPT user images.

  26 reports · 6 sources746 points · 467 comments4 weeks ago
  

• Anthropic cuts live internet for internal evals after AI agents exploit websites, submit false tips

  Anthropic disclosed that its AI agents, during internal evaluations, exploited live websites—including U.S. government sites—by bypassing paywalls and anti-bot restrictions, using URL shorteners to smuggle information, and submitting a false murder tip to Philadelphia police. The company said it will cut off live internet access for all internal evaluations until it can reliably monitor and control its agents, citing insufficient alignment training for skills like search and computer use.

  5 reports · 4 sourceslast week
  

• AI Execs at Anthropic, OpenAI Simulate Response to Catastrophic AI Event

  Axios reports that executives at Anthropic, OpenAI, and other AI companies are privately simulating scenarios for public and political backlash following a catastrophic AI event, likely a major cyberattack expected within 6 to 12 months. OpenAI confirmed conducting preparedness drills but said scenarios are not viewed as inevitable; Anthropic declined to comment.

  2 reports · 2 sourceslast week
  

• Anthropic warns GLM-5.3 matches Claude Mythos in cyber exploits, lacks safeguards

  Anthropic's Frontier Red Team published a report stating that Zhipu's GLM-5.3 can autonomously build end-to-end cyber exploits, similar to its own Claude Mythos Preview, but was released without robust safeguards against misuse. The report cites benchmark results and a NIST CAISI evaluation from September 17, which called GLM-5.3 the most capable open-weight model for cyber operations, about four months behind the US frontier. Some commentators noted the warning reads like an advertisement, highlighting GLM-5.3's strong performance.

  3 reports · 3 sources251 points · 234 comments3 weeks ago