Auth0 and Firebase both let you export user password hashes (though I believe you need to open a support ticket in order to do it in Auth0's case at least).
I think Cognito is actually one of the few with absolutely no path to achieving this.
Entra External ID does not allow export of password hashes.
It actually doesn’t allow a whole shitload of stuff that should be there by now.
For example, it supports passkeys, but has no UI for adding or managing passkeys. You have to build that yourself. In a separate app, on a separate domain.
I think its so that like people don't need to do an EPA assessment for their home backup generator right? They are very common where I live due to the instability of the power grid.
I took GP as asking for evidence that the reduced-cost Sol is actually a distillation of the previous-cost Sol. AFAIK, providers distilling or quantising models and offering them as the same model have not been proven.
I doubt he was claiming that. He's probably saying that the ability of Chinese companies to be able to distill frontier US models has put downwards pressure on the price of all models.
I'm saying that it is unclear that without distillation this wouldn't still be happening. There is a massive narrative that no one but OpenAI, Anthropic, and Google can make a model without distilling. But there's basically no evidence of that.
Alternatively, modern AI is good enough at optimizing its own kernels that it just keeps pushing costs down. Unlike the semi-decentralized inference provider community, OpenAI has both the talent and the compute to throw at the problem of making their models much more efficient to run.
GPU kernel optimization is just the kind of well-bounded problem with clear success criteria that AI loves.
To be precise, the distillation mentioned in this paper is not the distillation used by other model companies. In the one mentioned in the paper, your teacher and student model typically have similar architectures - and you typically need access to the full logits. What happens here instead is motivated by the fact that these companies don't have access to the training data and compute that anthropic/openai have. The distillation they do basically amounts to using traces from ant/oai models trained on much more data with a lot more compute (in many cases including the hidden intermediate tokens! turns out there were many ways to coax it out) and then either directly training on it or using it in many ways in post training pipelines. It falls under imitation learning, IMO.
And this is why he isn't being charged with destruction of evidence. He is being charged with destruction of property subject to a civil seizure.
The only reason they can charge him with this is that he destroyed it after being notified that it was being seized. If he has deleted it before then there would be nothing they could charge him with.
Right, but I think what makes this interesting is the destruction happened only after violating his civil rights. Had they simply seized the device the data wouldn't have been destroyed. He refused to give them the pass code several times before giving them the wiping code. He asked to leave and see a lawyer multiple times, which they refused to provide him with one. They held him, illegally, until he gave them his phone pass code.
And the whole reason they wanted in his phone (the real reason) was to figure out who he worked with in the stop cop city protest. That's why he was really targeted in the first place.
In cinema it works the same as in software engineering. There is an exception to normal copyright law where works made in the course of employment are treated as if the author is the employer rather than the actual author of the work. Its known in copyright law as a "work for hire" https://en.wikipedia.org/wiki/Work_for_hire
> In the United States, United Kingdom, and several other jurisdictions, if a work is created by an employee as part of their job duties, the employer is considered the legal author or first owner of copyright.
> It is an exception to the general rule that the person who actually creates a work is the legally-recognized author of that work.
Not paying anyone would be a violation of minimum wage laws. Or it would mean they are truly equal partners in the project who have no "employer" they are working under. In which case it does seem most fair that they would all own the copyrights to their contributions absent some other agreement to assign copyright.
> DRM exposes an API that user-space programs can use to send commands and data to the GPU and perform operations such as configuring the mode setting of the display.
> User-space programs can use the DRM API to command the GPU to do hardware-accelerated 3D rendering and video decoding, as well as GPGPU computing.
> Hugging Face alerted the FBI.
https://www.pbs.org/newshour/science/ai-agents-are-hacking-s...
reply