Sorry to hear. Just asking, why did you choose MIT license over something like GPL? It's not that they couldn't have copied you, but you would be in a different position vis-à-vis them now.

I would've chosen GPL personally, but a prototype repository already existed under MIT by the time I joined as a contributor through the fellowship. Alas.

It's actually fine to use MIT code and publish it under GPL, or to switch from MIT to GPL. You can still do that today on the repo, btw. It's the other direction that is thorny. Just fyi

Doesn’t matter. AI bros don’t care. AI wash it and release it closed. What are you going to do? Sue them?

...thus implying that you'd rather the preeminent safety-focused lab not have used your interpretibility research, as they're technically a for-profit company with various closed-source assets?

I am certainly biased by a passionate distaste for IP law in general, but I am honestly curious to hear the reasoning behind it in this case.

In case it helps: this is a "research prototype", so I'd assume that seeing one or more aspects used in a highly impactful paper would be about the best possible outcome. Is this about publish-or-perish stuff, perhaps (e.g. citation "importance" for the algos)? Or maybe you're okay with this in general, but are regretful because it's Anthropic in particular?

OP was pretty clear they wanted attribution, so you have hallucinated an entirely novel implicated desite for exclusion.

Seeing as the question posed out of "[honest curiosity]" to someone else is then immediately answered by yourself host without you even pausing to put on a sock, I think your honesty is very questionable.

Oh sorry, I must have misunderstood! My bad, honestly; I was under the impression that GPL vs. MIT was about copyleft stuff, which would seem to preclude Anthropic's use of that project altogether. Not 100% sure how it relates to attribution still but I'll have to look into it I guess.

I don't really understand what was "answered by [my]self host" nor what a "desite for exclusion" means but it seems beside the point I think. Also not sure what the alternative to curiosity is. Like, I'd be trying to entrap this prestigious researcher in some legal sense...?

Why can't Anthropic give credit though?

Who said they didn't? Not sure what "very little" attribution means, but presumably it doesn't mean "none."

I think I misunderstood GPL per a comment above so it's all kinda moot anyway, but my original understanding was that it was a copyleft license, so wishing it was GPL would be equivelant to wishing Anthropic had not used that work at all.

Attribution is certainly polite, and I sorta doubt Anthropic of all groups is intentionally witholding it for h-score or prestige, but that's all a much lower-stakes conversation than potential license violations. AFAIU.

If they are closed source and for profit, can they really be described as safety focused…

No, unfortunately. In this era of AI rewriting entire projects from scratch, copyright protection for software is effectively dead.

The question is less about protection than enforcement, I'd say. Courts may come to very different conclusions from what tech people would expect.

A small academic lab is going to sue a trillion dollar company with any sort of success?

Yeah, enforcement is the much bigger issue. If the IP got bought out by a giant lab, the situation could change quite substantially.

Protection and enforcement are kinda two sides of the same coin in most IP law. IP essentially grants you exclusivity for your works, and your protection comes from how well you can enforce that exclusivity. As an example, piracy exists largely because copyright holders simply can’t enforce their exclusive rights to distribution effectively on the vast, distributed Internet.

It’s still early days, true, and courts may find legal reasoning to address this... But I fear this stems from a fundamental yet little-known flaw in copyright for code, based on a legislative hack, which can only be made worse by caselaw. I ranted about it a few days ago: https://news.ycombinator.com/item?id=50320875

Regarding attribution, I guess he is one of the copyright holders, so his name should be included in the new license. Unless we are talking about MIT-0.

Im not an expert on these things, but I think there is another type of attribution, that I guess is what the commenter is more concerned about, and that is academic attribution. It is part of academic culture that you attribute significant influences in the articles you write. There are no licenses as such for this, as far as I know. You do it because that is what you do in academic circles. If you dont, you are not in the club.

Not sure if this is sufficient attribution, but TFA at least mentions VerInf:

> VerInf proved Llama 4 Maverick (the same 400B-parameter MoE model we prove) at ≈0.02 tok/s.

I wonder if this is partially letting agents.. agent, agentically and not know where ideas come form or how they come together, only whether it works or not.

The claude contributeth and the claude taketh away.

I’m so tired of hearing stories like this. The biggest most well funded corps in the history of mankind and all they do is steal the shit out of everything. It’s pathetic.

> MIT license

Live by the sword, die by the sword...

Nobody from Anthropic to comment on this?

I hope you don’t get that wrong, but maybe your best option is to see it positive. The top comment on a first page post on HN, gives you some publicity. Potentially more than if that would’ve not have happened.

You can also use this and create explanation video on it on Youtube, post on X, Bluesky and other in-detail stuff that only you and very little people are in the position to do.

I’m sure some are interested, and this could lead to a hire, too.

I know it’s always easy to say and I probably have less adversity faced as most people, but I really try to see them as energy enablers. Of the almost infinite options you have in life there is a big chunk of realities that are better than your current ones an adverse event can trigger emotions and therefore energy and push one to do things one wouldn’t have done in comfort.

Thanks for your work!

I agree with this. I often dedicate my life for months on end to prove novel and potentially valuable ideas just in the hope that one of the largest companies in the world can rip it off so I have the prestigious opportunity to make a youtube video about it. I can't believe that everyone doesn't do this.

Well I’m not saying to justify that, you can also start an activist movement to fight that, but I would rather move in the worlds that I can affect than in the one where I complain with no impact.

Guess I'm making a video about the system! My younger self would be proud about this... any tips?

I’ve only done a bunch of "thinking" videos. Nothing special, so might not be the best advice giver, but I’ll share what I learnt.

Don’t try to achieve any goal with it, except doing it. When doing it you’ll eventually get better in making videos.

The first video takes way longer than you think it takes, but the second probably takes half the time and it will be tremendously faster over time.

Try to reduce the friction on recording. My first training video on my app took me almost a full day to set up and plan with teleprompter and everything. After a couple of time I could do 3 in a row.

Making videos is still not "my thing" that I enjoy most about, but I’m still happy that I did it, if that’s what I need to I’ll do it. Right now coding has more leverage for me.

A technical write-up or a blog is an alternative. Or even reaching out to Labs. You could reach out to OpenAi/Anthropic/Mistral or your 'poison' of choice, or companies that do something similar. But this is now on your CV.

I guess if you’re “motivated enough”, you can [try to] put a positive spin on almost anything.

That improved my wellbeing

Sure like any other method akin to burying your head in the sand, this could be “helpful”.

I appreciate your positive energy here a lot. And I agree; all things can be dichotomies. This is how our entire team reacted to the news: it is good news and bad news both.

Claude being a official contributor here gives me the shivers. This feels like setting up a case for: We have the rights for everything you do with Claude. Or rather what they do with Claude for you.

I'm just annoyed that they treat it as a person. Claude isn't a contributor more than that shovel I used to dig a hole in the ground is. Even if it's a really good shovel.

In that case they should cite every author of every source used to train their model that enabled this result.

This is a big exaggeration. The commitment scheme is related but different (Ligero vs Zook). Matmuls are different (Freivald vs Sumcheck), lookups are different (LogUp vs Grand Product), the packing is different.

What original ideas are in VerInf that is also in this work besides the choice of benchmark?

I'll concede that "ripped everything" was written as a strong emotional reaction. But you are also exaggerating; there were a lot of novel ideas that we worked through used here. I'd point you to our choice of UD-Q4 and rules for how to convert to integers as a starting point.

You should have used Bedrock...no sharing of inputs with model providers ;-)