OpenAI to train on Daybreak Blue security-research sessions from Nov 29, with no opt-out
ccerrato147 · x · 2026-10-31
Security researchers shared an official OpenAI email: starting November 29, content from Daybreak Blue sessions on personal accounts may be used to improve OpenAI's models — even when the "improve the model for everyone" setting is off, with no opt-out. Business/enterprise customers and PSP/ZDR handling are unaffected.
@moyix confirmed it and called it "a horrible idea"; @seanhn argues it effectively steals security researchers' private work — something OpenAI promised not to do to math researchers — and will likely kill Daybreak usage by security researchers, underscoring the importance of open models.
Related event: OpenAI will train on Daybreak Blue chats from Nov 29 with no opt-out(5 posts)→
More from Companies & People
- Radar Lat podcast to break down Anthropic's S-1 filing as a business — yenkel · 2026-10-31
- Podcast: solving creator attribution and taste in the age of AI — hsu_byron · 2026-10-31
- Sam Altman says OpenAI is sitting on many unpublished mathematical discoveries — kimmonismus · 2026-10-31
- Sourcery founder interviewed 200+ founders in a year, landing Karp via Palantir coverage — eliano · 2026-10-31
- All-In breaks down Anthropic S-1 as 'greatest business ever' after Google sheds $500B in a day — markjeffrey · 2026-10-31
- Why is Zuckerberg the most hated figure in tech? Reddit thread probes the consensus — Dogbold · 2026-10-31