Fearing Data Misuse, Companies Turn to Open Models and 'Sovereign AI' Over OpenAI and Anthropic
fortune · reddit · 2026-10-06
Fortune reports that data privacy fears — that frontier labs may use customer IP to improve models or leak 'secret sauce' — are pushing companies to diversify AI vendors. Anthropic's June mandate of a 30-day chat retention policy on its most powerful models, Fable and Mythos, triggered an 'almost allergic reaction' among clients, despite zero-data-retention options. Security experts note subtler ways vendors can glean insights even without direct training. The result: rising corporate interest in 'sovereign AI' — owning the full AI stack from chips to proprietary cloud to self-hosted open-source models, a term previously associated mainly with governments.
More from Companies & People
- Google's Arohan: Gemini Nano chapter added to report, real inflection lands next year — _arohan_ · 2026-10-06
- Penn State IST department hiring multiple tenure-track AI faculty this round — windx0303 · 2026-10-06
- Coral-on-cream is now the most-copied AI palette: Claude, Mistral, Cursor, Replit, Cohere — TejasKumar_ · 2026-10-06
- Chick-fil-A says it will never use AI for drive-thru orders, sticking with humans — Polymarket · 2026-10-06
- Lumen Velar Returns to Courant3D With Four Talks on AI Filmmaking and Workflow Design — LudovicCreator · 2026-10-06
- latkins welcomes full-time joiner to Arcee AI — hardimanjames · 2026-10-06