GPT — Field Evidence
redditUnverified
GPT · GPT · incident
Field note
Potential poisoning of closed weights AI — I have been reading about the impending ban on Chinese Open Weight models here and it got me thinking about something insidious which could be at play soon. OpenAI/Anthropic could initiate a clandestine poisoning of Claude/gpt outputs to
collected 2026-07-23original 2026-07-23
Does this shift the US–China race?
Be the first to call it
Impact on the index
Regulation · major
US -110China +30
Directional contribution — before recency decay and per-type diminishing returns. How it works →
Related Front
US vs ChinaLikely
U.S. FrontiervsChina Open-Weight
Likely — capability gap narrowing on common tasks
Letters from the Front
Be the first to file a report.