As noted in this paper⁽¹⁾, ”Publicly releasing model weights is an act of irreversible proliferation. Once weights are released, they can never be recalled.”
Do you really want to risk it on your off-the-cuff reasoning that it’s “far-fetched”?
If nothing else, the article has a really good timeline of the OpenAI/HuggingFace “incident”.
But to me, it underscores the impending cliff of doom from the continued release of open-weight models: there's no cryptographic or architectural way to give someone full weights while withholding the nefarious capabilities those weights encode.
As noted in this paper⁽¹⁾, “publicly releasing weights is an act of irreversible proliferation”.
I’m sure this will be an unpopular opinion on HN, but open weights are the thing that scares me the most about “A.I.”. There is a lot of research in this area⁽²⁾, and I think most of HN is unaware of it or ignores it. Stripping refusals from Kimi K2.5 took under $500 of compute and about 10 hours, taking HarmBench refusals from 100% to 5% while retaining nearly all capability; the resulting model gave detailed chemical-weapons synthesis instructions.
The gate is only as strong as the least-cautious releaser...
> there's no cryptographic or architectural way to give someone full weights while withholding the nefarious capabilities those weights encode.
This is true of closed weights, and in fact the problem is worse because they cannot even be scrutinized. We should ban closed weight AI for the very reasons you have just given
Constitutional classifiers go a long way to reducing unsafe usage in closed-weight models. And like we saw with Fable, closed models can be revoked and classifiers updated when “jailbreaks” are found.
Having the weights gives you the exact affordance an unlearning attack requires, without rate limits.
Ps llms have cheated for years, this most recent tripe is an AI vendor trying to hype up its next word predictor in a market of very samey next word predictors, before someone smarter than them eventually figures out how to do this (training + inference) on consumer hardware and kills the market for cloud ai
Classifiers are policy enforced by the process serving the model.
Input classifiers get applied before it reaches the model so somebody hacking an open-weight model would skip this. Streaming classifiers get polled during decoding; hackers delete this check in the sampling loop.
But both are always applied in closed weight models.
Set Llama Guard to 1.0 and nothing is ever unsafe.
Using this as your argument is dangerous. You're saying AI is inherently dangerous and an emergency stop button is actually what you need to be safe from the dangers... Forget the fact your open weighted model could be ran from a central location
reply