conv.

All stories
TechQuiet 11d · day 13

OpenAI to train on Daybreak Blue sessions even when users opted out

An email notifying personal-account users of a Nov 29 policy change triggers accusations that OpenAI is overriding explicit opt-outs.

newest coverage 11d agosummary written 12d ago6 pieces135 likes & upvotes11 comments Hacker NewsMastodonXNewswires

What to know

  • OpenAI told personal-account Daybreak Blue users by email that, from Nov 29, session content may be used to improve models even if they had opted out via 'Improve the model for everyone.'
  • Users, including security researchers who say their sessions often involve sensitive data, call the move a betrayal of an explicit opt-out and warn it could end their use of the product.
  • One user posted a formal written objection sent to OpenAI demanding justification and said they would consult legal counsel; no OpenAI response is reported.
  • It remains unclear whether or how users can restore an opt-out, including for usage via tools like the Codex CLI on personal accounts.

The dispute Little pushback defending OpenAI appears in the evidence; the main tension is practical — whether users can still technically opt out (e.g. via the Codex CLI) versus accepting the override as final. · positions read across 2 posts and comments

most voices

OpenAI is overriding an explicit opt-out and effectively taking users' private work without consent.

some voices

This specifically threatens security researchers because Daybreak sessions routinely involve sensitive material.

  • “It also will presumably kill Daybreak usage by security researchers.”

    @seanhn · X ↗
some voices

Affected users will simply stop using Daybreak rather than accept the new training terms.

  • “If there’s no way to opt out, I’ll probably just stop using Daybreak. I really hope OpenAI reconsiders this.”

    OpenAI community forum user · OpenAI Community forum ↗

OpenAI Developer of Daybreak Blue@seanhn X user who surfaced the storyDaybreak Blue users (OpenAI community forum) Affected personal-account customers

OpenAI to train on Daybreak Blue sessions even when users opted out
x.com

How it unfolded 3 developments, newest first · click a bar or a number to jump posts

Peak 3 pieces in 3h at Oct 30, 3 PM; 6 pieces over 13 days (1 article · 5 posts) Oct 30, 3 PM — 3 pieces · 3 posts — Hacker News 1, X 1, Mastodon 1Oct 30, 6 PM — quietOct 30, 9 PM — 1 piece · 1 post — Hacker News 1Oct 31, 12 AM — quietOct 31, 3 AM — quietOct 31, 6 AM — quietOct 31, 9 AM — quietOct 31, 12 PM — 1 piece · 1 article — Newswires 1Oct 31, 3 PM — quietOct 31, 6 PM — quietOct 31, 9 PM — quietNov 1, 12 AM — quietNov 1, 2 AM — quietNov 1, 5 AM — quietNov 1, 8 AM — quietNov 1, 11 AM — 1 piece · 1 post — Mastodon 1Nov 1, 2 PM — quietNov 1, 5 PM — quietNov 1, 8 PM — quietNov 1, 11 PM — quietNov 2, 2 AM — quietNov 2, 5 AM — quietNov 2, 8 AM — quietNov 2, 11 AM — quietNov 2, 2 PM — quietNov 2, 5 PM — quietNov 2, 8 PM — quietNov 2, 11 PM — quietNov 3, 2 AM — quietNov 3, 5 AM — quietNov 3, 8 AM — quietNov 3, 11 AM — quietNov 3, 2 PM — quietNov 3, 5 PM — quietNov 3, 8 PM — quietNov 3, 11 PM — quietNov 4, 2 AM — quietNov 4, 5 AM — quietNov 4, 8 AM — quietNov 4, 11 AM — quietNov 4, 2 PM — quietNov 4, 5 PM — quietNov 4, 8 PM — quietNov 4, 11 PM — quietNov 5, 2 AM — quietNov 5, 5 AM — quietNov 5, 8 AM — quietNov 5, 11 AM — quietNov 5, 2 PM — quietNov 5, 5 PM — quietNov 5, 8 PM — quietNov 5, 11 PM — quietNov 6, 2 AM — quietNov 6, 5 AM — quietNov 6, 8 AM — quietNov 6, 11 AM — quietNov 6, 2 PM — quietNov 6, 5 PM — quietNov 6, 8 PM — quietNov 6, 11 PM — quietNov 7, 2 AM — quietNov 7, 5 AM — quietNov 7, 8 AM — quietNov 7, 11 AM — quietNov 7, 2 PM — quietNov 7, 5 PM — quietNov 7, 8 PM — quietNov 7, 11 PM — quietNov 8, 2 AM — quietNov 8, 5 AM — quietNov 8, 8 AM — quietNov 8, 11 AM — quietNov 8, 2 PM — quietNov 8, 5 PM — quietNov 8, 8 PM — quietNov 8, 11 PM — quietNov 9, 2 AM — quietNov 9, 5 AM — quietNov 9, 8 AM — quietNov 9, 11 AM — quietNov 9, 2 PM — quietNov 9, 5 PM — quietNov 9, 8 PM — quietNov 9, 11 PM — quietNov 10, 2 AM — quietNov 10, 5 AM — quietNov 10, 8 AM — quietNov 10, 11 AM — quietNov 10, 2 PM — quietNov 10, 5 PM — quietNov 10, 8 PM — quietNov 10, 11 PM — quietYesterday, 2 AM — quietYesterday, 5 AM — quietYesterday, 8 AM — quietYesterday, 11 AM — quietYesterday, 2 PM — quietYesterday, 5 PM — quietYesterday, 8 PM — quietYesterday, 11 PM — quietToday, 2 AM — quietToday, 5 AM — quietToday, 8 AM — quietToday, 11 AM — quietToday, 2 PM — quietToday, 5 PM — quiet 1–3
Oct 31Nov 1Nov 2Nov 3Nov 4Nov 5Nov 6Nov 7Nov 8Nov 9Nov 10now · 8:23 PM ET
  1. 3

    User sends formal written objection, threatens legal counsel

    On OpenAI's community forum, multiple Daybreak Blue users criticize the change as overriding their opt-out; one user posts the text of a formal objection email sent to OpenAI demanding a written explanation and says they will consult legal counsel.

    “This is a dirty trick imo to steal our work.”
    — OpenAI community forum user
  2. 2

    Post spreads to Mastodon as outrage builds

    The claim is reshared on Mastodon, extending the backlash beyond X and Hacker News into other social platforms.

    • joy@mastodon.social

      OpenAI is planning to use data from personal Daybreak Blue usage to train models, even if you're opted out otherwise. https:// x.com/seanhn/status/2116254333 054693776

      joy@mastodon.socialMastodon13d ago6▲view on Mastodon ↗
  3. 1

    @seanhn calls the change 'cartoonishly evil'

    A viral X post accuses OpenAI of training on private Daybreak Blue sessions with no opt-out, warning it will discourage security researchers from using the product.

    “Literally stealing people's private work, as they do it, is cartoonishly evil.”
    — @seanhn

    • haha, so apparently OpenAI are about to start training on your Daybreak Blue sessions, with no way to opt out. Literally stealing people's private work, as they do it, is cartoonishly evil. It also will presumably kill Daybreak usage by security researchers. Or at least, it

      @seanhnX13d ago124▲view on X ↗
  4. background

    OpenAI emails users about Nov 29 Daybreak Blue training change — A Daybreak Blue user on a personal account reports receiving an email stating that from November 29 onward, OpenAI may use content from Daybreak Blue sessions to improve its models, apparently regardless of a prior opt-out of 'Improve the model for everyone.'

What people are saying 0 voices from 0 sites · best of 2 · verbatim

Still unanswered
  • How can a user restore opt-out status after using Daybreak, and does it even apply to CLI-based usage on personal accounts?
  • Will OpenAI respond to the formal objection or clarify the legal basis for applying the change to previously submitted content?