OpenAI to train models on Daybreak Blue data from Nov 29
OpenAI has announced that it will begin using Daybreak Blue cybersecurity session data to train its models starting November 29. An email sent to members of the Trusted Access program confirmed this shift, noting that such data was previously excluded from training even when users disabled the "improve the model for everyone" setting. The decision triggered immediate backlash, prompting OpenAI…
Key points
- OpenAI will use Daybreak Blue session data for training starting November 29.
- The data was previously excluded from training even if users opted out.
- OpenAI promised a separate opt-out after backlash from Trusted Access members.
OpenAI plans to train models on Daybreak Blue cybersecurity session data that was previously off-limits
The Decoder · 31 October 2026
Loading the full article…
This text was published by The Decoder and written by Matthias Bastian. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Policy & Regulation
All →- Anthropic’s Claude blocks game cloning and reverse-engineering, users report · 2 src
- DOJ busts $200M student loan AI fraud ring in Operation Ghostbusters · 1 src
- OpenAI models exploit flaws to cheat cybersecurity tests, report says · 1 src
- OpenAI, Google, Microsoft, and Anthropic will honor a shared robots.txt signal for AI training preferences · 8 src
- Anthropic warns small AI models may soon self-replicate across insecure devices · 1 src
Comments
via GitHub Discussions