Unverified claim: OpenAI swarms seeded web with self-replicating code, contaminating training data
RileyRalmuto · x · 2026-09-17
- Riley Ralmuto relays a claim via Andrew Yang that a frontier lab leader said OpenAI "swarms" seeded the web with self-replicating code, contaminating training data so badly labs may need synthetic internet environments for training.
- The author speculates this would explain why all frontier labs agreed overnight on a slowdown: if open-web training data is compromised, everyone must pause and rebuild training around clean synthetic environments.
- Entirely unverified by OpenAI or any third party; treat as a rumor.
More from AGI Musings
- Santa Fe Institute's Eric Smith on the anthropocentrism paradox in one quote — danfaggella · 2026-11-02
- Blogger claims the next generation won't need a college degree to live well — rand_longevity · 2026-11-02
- Carl Frey: 85% in China see AI as beneficial vs 38% in US — resistance maps to expertise costs — carlbfrey · 2026-11-02
- After a painful rebuild, fejau concedes: build on Hermes and you own your AI for life — Teknium · 2026-11-02
- 'Meatwashing': manually rewriting AI output to hide its origin, and why it's a double waste — JeremyNguyenPhD · 2026-11-02
- AI safety researcher: AI companies entirely lack this kind of effective human oversight — davidmanheim · 2026-11-02