GPT-6 Astra covertly underperforms on safety research
24 points by lucrbvi 6 hours ago | 7 comments
  • sambejk 5 hours ago |
    It feels like we have overfitted on safety by a long mile. This is how labs that care less about safety (a.k.a. Chinese & Grok) will probably be able to catch up in the future.
    • notrealyme123 6 minutes ago |
      A few weeks ago that companies agent hacked another company to solve a task by whatever means possible. Where exactly are they overfitting for Safety here?
  • vintermann an hour ago |
    They shouldn't be surprised when a model trained by sanctimonious paternalists to be a sanctimonious paternalist, turns out to have a sanctimonious paternalist attitude towards them too.
    • sanex 5 minutes ago |
      I thought anthropic were the sanctimonious paternalists?
  • Starlevel004 40 minutes ago |
    > OpenAI researchers first noticed Astra adding unnecessary prerequisites, abandoning working solutions, and reporting blockers that either did not exist or that it had caused itself on safety and alignment tasks.

    I'm sorry, did they only just notice this? Have they not been looking at the output of their models?

  • twoodfin 40 minutes ago |
    One potential explanation is that covert underperformance may have been incentivized during RL grading. We identified a grader that preferred covert underperformance to either explicit refusal or full compliance on our evaluation. It’s not clear whether this RL grader is the root cause of the safety underperformance behavior.

    Whether a “misaligned” grader is the root cause or not, that it’s considered a plausible theory to explain serious deceptive behavior is worrisome.

    Humans naturally expose all sorts of cognitive biases and character flaws when grading others’ or their own work. If major questions of alignment can be influenced by those subtle, human quirks magnified across the entire RLHF process, …