Anthropic · The Guardian Technology
Anthropic investigates report of rogue access to hack-enabling Mythos AI
Compiled by KHAO Editorial — aggregated from 1 outlet. See llms.txt for citation guidance.
◌ Single Source
The AI developer Anthropic has confirmed it is investigating a report that unauthorised users have gained access to its Mythos model, which it has warned poses risks to cybersecurity.
Key facts
- Mythos was the first AI model to successfully complete a 32-step simulation of a cyber-attack created by AISI, solving the challenge in three out of its 10 attempts
- Kanishka Narayan, the UK’s AI minister, has said UK businesses “should be worried” about the model’s ability to spot flaws in IT systems – which hackers could then act upon
- The model has been vetted by the world’s leading safety authority for the technology, the UK’s AI Security Institute (AISI), which warned last week that Mythos was a “step up” from previous models
- AISI said Mythos could carry out attacks that required multiple actions and discover weaknesses in IT systems without human intervention
Summary
The US startup made the statement after on Wednesday that a small group of people had accessed the model, which has not been released to the public because of its ability to enable cyber-attacks. “We’re investigating a report claiming unauthorised access to Claude Mythos Preview through one of our third-party vendor environments,” said Anthropic. A “handful” of users in a private online forum gained access to Mythos on the same day Anthropic said it was being released to several companies including Apple and Goldman Sachs for testing purposes. It reported that the unnamed users got to Mythos through access that one of them had as a worker at a third-party contractor for Anthropic and by deploying methods used by cybersecurity researchers.