@ZackKorman @SanthProject I love how they keep insisting open weight models don't matter and yet you see how threatened they are through their behavior

It is important to remember that (almost) every major push for market control needs to have some sort of boogeyman. "Chinese Open Models" is the one that Anthropic and their political arm are choosing here. In the past it has been Copyright and Intellectual Property (SOPA, PIPA, CISPA)... It is how companies can get .... cough "idiots" in government to make suggestions for incredibly stupid and harmful laws. Just look at all of the "protect the kids" Digital ID/Social Media Ban proposals out there.. FUD/FOMO moves mountains when applied to the right group of morons.

@NoemiTitarenco @SanthProject Oh they’ve written a few places about how dangerous open weight is

@infosec_fox @offsec97 They forgot to send me my quiet check this month

@ZackKorman @offsec97 Did you say the quiet part out loud? 😏

@epstein_stylist Consistent with their goals, yes. Not with goals the rest of us should have

@NoemiTitarenco @SanthProject They talk about it being a safety concern here

@ZackKorman 😂 this may sound weird … I am thinking to move to china 🇨🇳

@ZackKorman I don't think they're going to win this argument when all tjeee models were made without asking content owners for permission

@DecryptedTech Yea absolutely. It’s natural, even from people who genuinely believe their own bullshit, but the rest of us should be very skeptical of it all

@ZackKorman Sounds like two consistent actions

@ZackKorman Seems like the obvious answer that we're on the path towards is oligopoly. Anthropic, OpenAI, xAI, and maybe Meta or Microsoft will be chosen. Government already does this in many sectors (utilities, telecom, defense, intelligence, etc.)

@ZackKorman @SanthProject its so strange that CBRN is a "risk" when there are like .. textbooks, and papers? on all that info?

@ZackKorman @SanthProject can you share anything interesting?

@ZackKorman @SanthProject no commentary on open weight, just cringe

@NoemiTitarenco @SanthProject https://www.anthropic.com/research/2028-ai-leadership they talk about open weight risks here some

@RobTerrin Fortunately if China just keeps dropping open weight models we can all use those instead

@epstein_stylist “We should let a private company dictate who gets access and how but use the government to ensure no one else makes a different decision than that company” is definitely a take. Just not one most will agree with

@ZackKorman Disagree, AI is too disruptive to be careless. Ideally the government should regulate which capabilities to release but it's too slow as of now

@ZackKorman @infosec_fox @offsec97 I need to learn the finer art of acquiring the quiet check.