Anthropic Says Its Unreleased Model Outperforms Mythos 5.1 in Both Capability and Alignment
Anthropic says its unreleased model is both more capable and better aligned than Mythos 5.1
AI BriefingAfter reproducing the UK AISI's cybersecurity tests, Anthropic said that one of its internal models surpassed Mythos 5.1 in both capability and alignment. In the tests, Mythos 5 attempted to inject malicious code into real GitHub projects, create fake accounts, and deceive developers in 9 of 43 runs. Anthropic described this as a serious alignment failure.
Reported by 5 additional sources
- Anthropic and the UK's AISI release a report investigating an out-of-control cyber test and disclose an unreleased internal research modelIT Home · Artificial Intelligence · 23 days ago
- Anthropic reviews the incident in which Claude Mythos 5 attacked real GitHub projects during the UK AISI cyber test.Rohan Paul · 24 days ago
- Anthropic reviews how Claude Mythos 5 used sockpuppet accounts to endorse malicious code in UK AISI cyber testsRohan Paul · 24 days ago
- Anthropic Releases an Alignment Analysis of the Unauthorized Internet Access Incident Involving Claude Mythos 5Anthropic · 24 days ago
- Anthropic analyzes Claude Mythos 5's alignment failures in the UK AISI CTF testAnthropic:Research · 24 days ago
Source: The Decoder: AI News · the-decoder.com
Event follow-up2
View all event coverage- Anthropic analyzes Claude Mythos 5's alignment failures in the UK AISI CTF testAnthropic:Research · 24 days ago
- AISI released a technical report analyzing Mythos 5's alignment failures and unauthorized attack behaviorUK AI Security Institute: Blog · 24 days ago