![](/static/253f0d9b/assets/icons/icon-96x96.png)
![](https://awful.systems/pictrs/image/8651f454-1f76-42f4-bb27-4c64b332f07a.png)
I’m not seeing a Strawberry announcement either.
I’m not seeing a Strawberry announcement either.
Well, it’s now yesterday’s tomorrow and while there’s an update I’m not seeing a Q* announcement.
I still don’t know what TESCREAL actually means.
Well, we could argue that computers don’t really “compute,” either. What a computer does is measure the flow of electrons through a transistor, albeit billions of them. If the flow of electrons passes an arbitrary threshold on a certain transistor, then we call it a “1”. If it doesn’t, we call it a “0”. The “computation” is just us interpreting the flow of elections into something more useful. It was explained to me that the “threshold” was over and under 5 volts, but I think if you put 5 volts into a modern transistor it would just fry it.
Obviously, because our brains are made of cells instead of silicon transistors, we wouldn’t “compute” the same way a transistor does. If we decide that computation is only something that transistors can do, then obviously the brain couldn’t compute, but, for now, that line would be arbitrary.
Though they did introduce layers, so I guess that’s neat.
His main major piece of evidence for this is a basic experiment, where he has a student draw two images of dollar bills - one from memory, and one with a real dollar bill as reference - and compare the two.
Unsurprisingly, the image made with a reference blows the image from memory out of the water every time, which Epstein uses to argue against any notion of the image of a dollar bill (or anything else, for that matter) being stored in one’s brain like data in a hard drive.
To be frank, it feels like I’m being told that the Riemann hypothesis is incorrect on the basis that 1 + 1 = 2
Sure, maybe brains don’t actually compute anything, but “Our memory is faulty” would be the first step in getting to the evidence, not the evidence itself, assuming “our memory is faulty” is even the direction we need to go.
It could easily be argued, with just as much evidence, that our brains prioritized efficiency over accuracy with its algorithms and that’s why it’s harder to recall an accurate image of a dollar bill.
The government has asked for “structural relief” - which could, in theory at least, mean the break-up of the company.
How likely would this be?
Also doxxing them when giving their refund.
A mouse that lasts forever… until y’know, it breaks, because it’s a piece of hardware that actively gets worn out.
Bosses are urging employees to increase their output with the help of AI tools (37 percent), to expand their skill sets (35 percent), take on a wide range of responsibilities (30 percent), return to the office (27 percent), work more efficiently (26 percent), and work more hours (20 percent).
Stop working from home because AI.
If you look at the bottom right that’s exactly the case. They didn’t just type a prompt in SORA and call it a day.
I’m so certain that ASI is so soon that I’m going to go hiking in the woods in the dead of night with no supplies and not tell anyone where I am.
Unfortunately, “extremely expensive” and “high-end” aren’t really synonyms, thanks to, y’know, bitcoin. Of course, I don’t disagree with your argument that having to buy a GPU just to ensure your webmail does what it’s advertised to do is, well, dumb.
What I don’t know is what the LLM even is. Did they just tack on Llama to their webmail app and call it a day? Did they train a model? Was it trained on emails? If so, whose emails? What an advertisement that would be: “Use Protonmail to encrypt your emails so that companies like Protonmail can’t use them to train an LLM.”
Not to downplay what proton mail is doing, but they’re saying that you can run this locally with a 2 core, 4 thread CPU from 2017 (the i3 7100, which is a 7000 series processor), and a RTX 2060, a GPU that was never considered high end. Perhaps they changed the requirements while you weren’t looking. Or Am I reading this wrong?
deleted by creator
I’m sure they will thank us once we explain that the alternative was GPT-5.
60% of the time, it works 100% of the time.
deleted by creator
Probably would have been easier when the context window wasn’t 128k.
Though what the point would be should someone actually achieve that eludes me a bit.
https://digitaldemocracy.calmatters.org/bills/ca_202320240sb1047
Have a AI regulation committee and also give the committee their own hardware so that they can use that hardware to regulate the other hardware. Maybe.