Another potential takeaway is that the models all gathering around the same point supports the idea that there is a ceiling to LLM capability.
This isn't an anti-American sentiment. It is an anti-corporate/regulatory capture/embrace and extinguish sentiment (which probably reads the same to many people these days).
But they didn't find it. The Big LLM provider accepted guilt and paid a fine.
You can argue whether it was a fair amount they paid, but there is no legal precedent that was set. It's still considered theft.
Training from copies has been ruled fair use because it's "transformative" and not simply "derivative."
This is obviously debatable, but that's where the debate is at the moment.
Because of the rulings of a couple of judges. Is that actually what the majority of people think?
> Copyright law only considers illegal ownership of a work
That's definitely not true. File sharing, for example, is illegal even if you legally own the original copy you're sharing.
Similarly, copyright has something to say if I read a legal copy of harry potter and then create a new work in that world.
There's a good reason for the law not to be based on what the majority thinks.
Sure i don’t think the majority get to dictate things like who has rights or who the law applies to. That doesn’t apply here tho.
Because that use case is actually permitted by law.
The law was written before the idea of an LLM existed, and some judges in some specific cases decided the previous law covered this usage.
So, it comes down to if you believe a couple judges ruling on a couple cases is the right way to determine a world-altering new legal framework.
It's not an either-or.
That's not how it works. You have to give it back.
Otherwise, the distiller can just pay a fine (no larger than the original did) and be okay then, right ?
Not sure what part of being charged guilty and paying a fine you see as "free".
Where the new generation of LLMs (Fable, Sol) shines is tasks that are much harder than typical soft eng, yet that still have a verifiable answer, think mathematical proofs or exploits. I think there's still a good amount of low-hanging fruit in those (and similar) areas.
The next frontier after that is tasks that don't have automatically-verifiable answers, and may not even have correct and incorrect ones in the strictest sense of the word.
Reasonable lawyers might disagree on the question of "which trial strategy do I use given the following set of facts." There are answers that are clearly wrong, but being able to choose between many plausibly-correct ones requires many years of lawyering and seeing many trials play out. I do suspect that most lawyers are far below the ceiling that a hypothetical immortal lawyer that has practiced for an infinite amount of time would have achieved.
So instead of relying heavily on human bottlenecks, you focus on agentic task verification since that's the low hanging fruit and verifiable at scale?
LLMs are certainly more knowledgable, but maybe not more intelligent, arguably. It's possible we're approacing a ceiling indeed.
Model capability might be on an asymptote appraching but never quite reaching parity with human intelligence.
Many extended kinds of verification can be done by LLMs, but they need to be able to follow instructions reliably and agentic task orchestration may be critical to that verification process.
There is no doubt they will surpass us as there is a lot of easy to reason about information that they can verify as incrementally proven by other knowledge. The trick is knowing what can be proven with existing knowledge and what needs human evaluation.