This is going in Metacurity's long reads issue on Saturday, but I urge folks to read it now.

[metacurity@infosec.exchange]

This is going in Metacurity's long reads issue on Saturday, but I urge folks to read it now. Really good description of how Anthropic figured out the power of Mythos.

How Anthropic Learned Mythos Was Too Dangerous for the Wild
https://www.bloomberg.com/news/features/2026-04-16/how-anthropic-discovered-mythos-ai-was-too-dangerous-for-release?embedded-checkout=true