That doesn't happen much, not unless the worker is forced to go back because of crappy H1B policies. The fact is that nearly all foreign workers would love to stay permanently.

Well, there is a non-zero probability that the Earth could enter a runaway warming cycle and become another Venus, with surface temperatures exceeding 400C. The planet wouldn't be destroyed, but life would probably be wiped out. AFAIK we haven't found any life form that can survive above about 120C.

That seems unlikely, though, given that Earth's hottest phases so far (after the crust cooled, anyway) have been considerably cooler than that.

So... remain in Mexico is the only one? Mexico has said they won't agree to reinstate it, and without their cooperation it would put the US in violation of our treaties. The president can't legally break treaties; Congress needs to act for that.

I hear you. This is Google, though, and they already do automated transcripts of Meet meetings and YouTube videos. Odds are good that you'll be able to easily get a skimmable and searchable transcript, and to be able to jump into a location of the video by clicking on the transcript. Might not be there in the first release, though.

We'll see, but I suspect that integration into Google's other office tools may give it a solid place. It doesn't have to be the best, it just has to be good enough, and well-integrated.

LLMs can't but there are theorem proving systems that can. Of course, this is just another example of how LLM's -- whatever their strengths -- are not general intelligences. We have many domain-specific systems that meet or exceed human capabilities, but we don't have artificial general intelligence at all, and don't know how to get there. I'm sure we will get there because I don't believe there's anything supernatural about human intelligence, but we don't know what we don't know about general intelligence.

True, if by "AI" you mean Artificial General Intelligence.

And that's what it will be when it is AI, too. It's almost certainly not the composition that's the problem, but the structure. We don't yet know how to structure the words, loops and integers to create AI. We will. But no one knows when.

I have, been using it since launch. Well since the opening of the general beta; I didn't try to get into the limited betas.

No, you wouldn't. Not after a few hours of experience with it. You pretty quickly learn what sorts of situations it handles well and which it might need you to take over. This doesn't mean you can go to sleep or stop paying attention to the road, but it does mean that you can relax more than if you were driving yourself. I find that my focus changes; I stop paying a lot of attention to the small details and instead pay more attention to the big picture. I feel like I have a much better sense of what's going on all around me when I'm using FSD -- and driving with FSD is less tiring, which becomes very apparent on long road trips. I dislike driving more than a couple of hours without it.

I agree with your argument about the semantics of the name, but what it's called has no bearing on whether or not it's useful.

Actually, we now know nothing more about whether it will or will not happen than we did before Elon said it.

In order for Elon's statement to assure us that it will not, Elon would actually have to know something about it. He doesn't. No one does.

That's not an example of a corporation being successfully sued by shareholders because they failed to prioritize profit over safety.

Is air safety decreasing? It doesn't appear to be: https://www.airlines.org/datas...

You're mixing pensions with 401ks, and they're entirely different things.

Pension managers have specific payouts they need to meet, on particular schedules. They might well have reasons to browbeat corporations into doing things.

401k managers don't have any of that. The money they manage is owned by the individuals, there are no specific payouts planned or required, and even how the money is invested is generally directed by the individuals. Of course, most individuals put their money in mutual funds or ETFs, from among the offerings provided by the brokerage. The managers of those funds may have some incentive to manipulate boards to goose their short term returns, but not much because they typically include a broad basket of securities and long-term funds need to have good long-term performance. They could goose the short term returns from a company and then cash out of that position before the damage shows up, but if they do that very much then companies will have strong reasons not to cave into their demands since doing so will hurt the company in both the medium term (when the fund cashes out) and in the long term (when the chickens come home to roost).

I don't think 401ks have anything to do with this sort of short-sightedness. Executive bonuses are a much more plausible explanation.

They're not required to. Find me even one example of a corporation being successfully sued by shareholders because they failed to prioritize profit over safety.

Yes, officers of corporations have a fiduciary responsibility to maximize shareholder value, i.e. profit (assuming that's what their articles of incorporation say, which is normal), but it's very, very hard for shareholders to hold them to it unless they're doing a really bad job. And with something like safety, especially safety of customers' passengers, any lawsuit that argued that prioritizing safety failed the fiduciary responsibility would be extremely easy to counter. The CEO would just get on the witness stand and say "Well, we knew that safety problems would cause airlines to stop buying our products, and could also create huge legal liabilities. We judged that prioritizing safety was the most prudent way to maximize shareholder value."