Stella Biderman and Aidan Clark debate open-source safety disclosures
Clark warns public safety methods are easier to bypass.
I somewhat regret the language here. Releasing GPT-OSS and not telling people what they did does not make the world worse than doing nothing. But their refusal to engage with the open source community on how to make models safer makes the world a much more dangerous place.
Combined views
643
7 posts, first seen 3h ago
Stella Biderman and Aidan Clark debate open-source safety disclosures
Clark warns public safety methods are easier to bypass.
I somewhat regret the language here. Releasing GPT-OSS and not telling people what they did does not make the world worse than doing nothing. But their refusal to engage with the open source community on how to make models safer makes the world a much more dangerous place.
Reactions from ranked influencers
6 posts@_aidan_clark_ @BhaskarSteve I don’t think “decreasing the search space” makes sense as an explanation because it has no connection to how people actually make models more dangerous: direct finetuning on hazardous capabilities. I also don’t believe that security by obscurity is real security at all.
@_aidan_clark_ @BhaskarSteve See also this clarification:
I somewhat regret the language here. Releasing GPT-OSS and not telling people what they did does not make the world worse than doing nothing. But their refusal to engage with the open source community on how to make models safer makes the world a much more dangerous place.
@BlancheMinerva @BhaskarSteve For what it’s worth — and to clarify this is not something I feel I can do unilaterally — I think we would be pretty excited about partnerships which let us safety-harden external models (including OSS ones), just less excited about sharing the techniques themselves!
@BlancheMinerva @BhaskarSteve The core reason because we do think knowing a technique makes that technique easier to undo (I’m very confused by the claim that is not obviously true, I suspect in different regimes you would immediately agree?)
@_aidan_clark_ @BhaskarSteve We have reached out about this in the past and gotten no response. Do you have a sense of who would be a good person to talk to?
@_aidan_clark_ @BhaskarSteve I don’t think “decreasing the search space” makes sense as an explanation because it has no connection to how people actually make models more dangerous: direct finetuning on hazardous capabilities. I also don’t believe that security by obscurity is real security at all.