Signs that the bubble is about to burst. They are now claiming AGI despite being fundamentally away from it. An LLM can not ever achieve AGI. It may reasonably well PRETEND to be AGI, but at the core it's just a highly successfull letter-prediction machine. Or as a metaphor: It is "intelligence" in the sense that a toddler carefully placing one foot in front of the other while trying hard to not fall over is "running". Nope, isn't and never will be. You need a qualitative change to get there.

If they now CLAIM that they've achieved AGI, it's a sign we're nearing the end of LLM development and they're getting desperate selling the latest incremental updates. Which means the illusion is about to come crashing down.

Which is good news. Fewer data centers and lower memory prices will benefit us all.

What are you ranting about? If you know confidential information from your employer, you are NOT allowed to profit off it. You are also not allowed to publish it, etc. That is what the law says. This has nothing to do with capitalism.

A really dumb lie by misdirection. Natural climate change happens massively slower and things can adapt.

Exactly. But lying, even directly, seems to be getting them business because people are stupid.

That is not the discussion. The discussion is whether the tool should be created and made available. There are lots of tools that are expert or special permission only or totally banned.

That problem is multi-dimensional. The raw number is a useless metric.

Just to elaborate a bit in case you really do not understand, humans typically are only somewhat off in a job they know how to do. And if they are more off or often off, they get fired. LLMs are sometimes extremely off _and_ you cannot fire them.

Do you really not understand how the requirement of keeping your employer's secrets and not profiting off them works? Seriously?

Yes. But since these models cause brain-rot, the very skill needed to understand that is being attacked. Probably planned like that from the very start too.

There are a bunch of inductive algorithms. x_n+1 = x_n + f(x_n) dt is an integral curve. Doing it discretely is a a piecewise approximation of that curve. The vast majority of neural nets we use today are piecewise linear function approximators.

I see what you mean, but heck, that's not even playing every day - More like every other day on average, and I often burn a few hours on Sundays.

But that 4% stat is interesting - What this tells me is Game Pass is effectively a gym membership where the entire business model depends on 95% of people never using it.

Not at 15 hours a month it's not. I'm a total casual, and even I'd burn through that in under two weeks.

This must be some form of Anchoring. They fully plan to come back in a week or two with 10x the hours, hoping everyone that would have laughed even at that, initially, suddenly decides it's not so bad. If not, this is Microsoft announcing the end of Game Pass.

That he did the trade is already a crime, whether the nature of the trade adds a crime or not.

Indeed.

You sound like you are massively overdoing some drugs. Or maybe generally psychotic? In any case, seek help, urgently.

May have something to do with their abysmally bad business situation. Obviously, the usual idiots will eat this up and believe every word...