lynndotpy

I'm not asking this in rhetoric; do you happen to have a link to this comment? I ask because a comment like that would be something of an insane thing for someone to say.

For those who can't see the video, I think the above two commenters are being facetious for humor.

It is funny because the video is 50 minutes long which requires a difficult soldering job and chip reprogramming. This does not include the process of opening the Macbook Pro, disassembling it, removing the logic board, nor the reverse process.

Did you know Moore had two not-as-famous cousins with their own laws? Sam and Leslie.

Sam predicted price-per-transistor-per-square-inch density would asymptote, while Leslie predicted it would level off and then regress.

Leslie's Law is in full effect, it seems.

Yeah, I was wrong. Page two too, I commented too soon.

NP is not yet proven to be subexponential but still, ETH disproven is huge

If I am understanding this correctly (and I am still reading this!) and if this is true, this means NP is sub-exponential. Sub-exponential does not mean P=NP. But yes, this is huge if true.

This means hard problems aren't as hard as we thought, and it's still possible P=NP. If it turns out to be true P=NP, we're a little closer to that. (I hope P!=NP!)

Every paper which assumes ETH is true might now have their conclusions be disproven.

I would be surprised if >10% of people didn't notice a difference between 60Hz and 120Hz.

Whether the hardware can keep up with Liquid Glass and push 120 frames a second is a different story, of course.

This isn't surprising but it's good to have reporting to point to. The only way a general purpose LLM like these will match the performance of a bespoke classifier is if it has access to such a model.

I parsed it as "A record that is ((160 bytes and an allocation) in C++) is ((29 bytes and none) in Goose)."

So, I read it as, "C++ uses 160 bytes and an allocation, when Goose requires 29 bytes and no allocation".

I would also like a source for this, I do not believe it is true and founded that "an LLM has a latent concept of its own confidence that is internally consistent".

LLMs generally cannot report the outputs of their own activation functions. Taking the output of softmax as if it were a probability score is considered a first-year PhD student misconception.

You are only talking about what you're feeling. Care to actually make a statement about what you think is true, besides implicitly denigrating someone else?

Still bleary eyed, but: Most severe ones are in SMB and KSMBD, which apparently have implementations in kernel space. I think this doesn't impact anyone not using KMB.

Others are LPEs, but two have been deleted by now, so can't be sure of the details.

TrendMicro's site opens up with a big big "Advancing AI-Native Security with Anthropic". Which checks out I guess.

I am not affiliated with PHP in any way or sense, I am saying this as someone who is able to see PHP's announcements, and there was no patch or mitigation announced by PHP.

I don't think pulling mitigations from random users online during a developing situation is a good idea. I think shutting down ones PHP services for the time being is a very good idea if that's an option.

I do not have details about any mitigation. There are rumors that people have independently come across php2shell exploits.

I shut down my servers five hours ago. This might have seemed premature at the time.

I think the best course of advice for most people is to shut down their servers as soon as they possibly can.