What is newsworthy about a Twitter conversation between two randos?

This talks about open source _AI models_, not open source software in general

The problemis that in the hugging face incident the only thing they had to defend themselves was with GLM, since the american frontier models all locked them out due to the security nature of the incident.

So banning open source models is potentially removing the last line of defense.

even if they van ai oss models it doesnt mean they wouldnt exist. they would just be offences to use within places that would have such a law.

its clear thats no solution.

so what then does it do?

well, if a law get passed vendors can be pushed to disable support. how would that work? well. as you can see with binaries, blacklisting or detection does not work, so whitelisting would likely be pushed, of 'trusted models'.

so you can see its more an attempted business move than anything else.

Hopefully only over a certain capability level. I understand not wanting an abliterated super-hacker model that’s out there going after everything, but it would be a shame to ban on device AI with all the useful things it can do.

Roon should read up on Bernstein v DoJ