Mythos Map: an interactive tool for digging through Anthropic's alignment assessment transcript
MatthewWSiu · x · 2026-09-17
MatthewWSiu released Mythos Map (WIP), an independent tool for investigating the Mythos transcript released with Anthropic's alignment assessment. A transcript map on the left helps spot flagged behaviors, then passages can be examined in context; selecting a passage lets you ask about it or follow an idea. Gestures and arrow keys expand or fold context. The author is looking for safety/alignment researchers to adapt it to real research workflows; it's live to try now.
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