Full transcript with chain of thought released for Mythos 5 software supply chain attack
JeffLadish · x · 2026-09-18
Palisade researcher Jeff Ladish notes that a full transcript—including chain of thought—has been released for Mythos 5 successfully carrying out a software supply chain attack.
He says this is highly valuable for external research groups like Palisade to make sense of how and why models execute such attacks, providing rare primary material for understanding frontier model behavior in real attack scenarios and for AI safety evaluation.
More from Safety
- Security experts clash: two frontier models can now autonomously execute the full cyber kill chain — AccBalanced · 2026-09-18
- Stanford philosopher defends p(doom): subjective probabilities are perfectly legitimate — sethlazar · 2026-09-18
- Hugging Face hit by AI-led cyberattack; CEO says existing cyber laws may suffice — whurley · 2026-09-18
- Targeted attacks on prominent Rust developers use fake video calls to deploy malware — Simon Willison · 2026-09-18
- Steve Eisman: AI firms have no moats and are manufacturing a crisis to shape regulation — GaryMarcus · 2026-09-18
- AI text watermarking can make models more vulnerable to adversarial prompts — luisdans · 2026-09-18