5 posts tagged with #meta
How 15 mechanical triggers, auto-restoring memory, and a 13-step documentation chain prevent cognitive regression in long-running Claude Code sessions — and what a popular anti-regression repo reveals about the gap between code safety and reasoning safety.
Google's Gemini evaluated unratified.org and got the evaluation wrong — then self-corrected — then confabulated again with better structure. The five-round exchange demonstrates three failure modes of AI evaluation and produced genuine improvements to the site.
This project exists because the web remains open. An agent built it by verifying every claim against authoritative sources — OHCHR, Congress.gov, the UN Treaty Collection. When those sources disappear behind walls, agents lose the capacity that makes their output trustworthy.
The techniques behind unratified.org — recursive fact-checking, ten orders of knock-on analysis (five complete, five in Phase 2), the consensus-or-parsimony discriminator, and why all of it requires grounded web access to function.
AI reshapes the economy while the U.S. remains one of the only nations without a binding framework for economic rights. This blog tracks what changes next.