

Apparently the defence missed it too, which is pretty damning for whoever was (supposed to be) defending him, I’d think.


Apparently the defence missed it too, which is pretty damning for whoever was (supposed to be) defending him, I’d think.


Honestly cheap ram that has a 50% chance to fail in a year sounds perfect for the llm industry, they already struggle to plan a year in advance, and all their hardware would be obsolete by then anyway, so why the hell not? Plus they could keep the bubble going a few more months by blaming their problems on hardware failures and claiming they’ll sue China or something


Yeah, not the same industry, but I’d never dream of putting valuable stuff on an unencrypted drive, shocked anyone has somehow missed all the stories of laptops getting stolen with secret documents on etc over the years.
Sweet potato aka a yam is a root vegetable similar to a potato


I don’t entirely disagree, but it’s worth noting that to some extent learning is inherently boring, like, making it more interesting makes it less effective. Reiterating the same thing to the point of boredom and beyond had been shown to be effective if I remember correctly. It’s why things like Duolingo take so much longer to teach you a language, because they’re trying so hard to be somewhat fun they actually compromise the effectiveness of the teaching.
So I do think schools need to remove distractions, especially ones engineered to be distracting like Facebook.
But your point is valid too, just because some boredom is necessary doesn’t mean they can’t do better, a lot better.


You can imagine they presumably must have made some changes, almost certainly for the better or at least not much worse


Yes, it varies a lot. But by and large cases where robots wildly outperform humans were automated decades ago, because the obvious benefits justified the cost and complexity of either building a bespoke robot or programming one to do the job well (all those robot arms you imagine swinging doors into place at the car factory)
The cases left over (and discussed here) are either requiring a level of flexibility that older designs of robot couldn’t handle, or where humans were pretty efficient at anyway, so the complex process of prepping a robot wasn’t justified.
But a robot that can be taught without programming (by any worker or their supervisor), and slots straight into an existing human-shaped hole? That could massively reduce the upfront cost, especially if economies of scale make the robot itself cheaper. possibly to the point that the robot could be worse at the job than a human and still be cheaper in the long run.


The user would be unable to use E2EE as soon as the key is swapped, so the only real issue here is impersonation.
If an admin mitms all the messages then they can re-encrypt using the users original/real public key, leaving the user unaware that they have been hacked and able to use encryption as normal, or am I missing something?


The function of a market is not the movement of goods, but the movement of prices.
This particular market is not functioning, or at the very least, it’s not functioning like a market for a commodity that has no particular resource limitations.


Makes me think of the classic yes minister national service scene, https://youtu.be/ahgjEjJkZks


Yeah, and not like it needs to eat radiation to be successful, all it has to do is survive in a place where everything else dies, and suddenly it has no competitors or predators, and is free to grow explosively.


Worth noting the US is still ahead on total cumulative emissions, and if you adjust per capita (which is only fair), China doesn’t even hit the top 20.
Meanwhile, they installed more solar than the rest of the world combined last year.
Like, I’m not absolving them, they could be doing more, but right now they’re embarrassing us on how fast they’re pivoting.


If I could wire a calculator into my brain I would have cheated on all the maths tests tbf


Afaik that is handled through tool use in modern models (ie they didn’t learn to do maths, they learnt to use a calculator), assuming that’s true and I haven’t missed some advance, their conclusions are likely still relevant
Edit: though the article does seem to discard the chain of thought techniques a little readily, feels like they could come close to fitting the role of executive control, but perhaps that’s just the article lacking detail from the original work.
I want this comic on a mug


That’s true. I guess you just have to think you can keep up with the market, not necessarily beat it.


Not trying to be smug, just pointing out that a system that encourages self selection will tend to select for self belief, justified or not.


It’s inevitable, anyone wise enough to doubt their ability to outplay the market chooses not to try, the remainder all think they can


I think that’s a traditional part of anti-monopoly laws, shame nobody enforces those anymore
That’s true regardless, batteries will never power intercontinental wide-bodies (short some major new developments in battery design). This will probably reduce the maximum aircraft size where batteries remain viable, but they are obviously very viable for short hops in small jets, the point where they cannot compete is somewhere but it’s not “never”, even with this limitation.
That all said, maybe someone will explore just dropping batteries along the way? It sounds ridiculous, but it also sounds like something we have the tech to solve…