You left out the robots. Enslaved to do what when the robots are cheaper?

Be careful there. Lots of AI is being put to silly, useless, or unreasonable uses. OTOH, lots of it is being put to extremely productive uses. (OK, 20% improvement in output, but also an increase in expenses.)

ISTM, that PART of the AI hoopla is a bubble. Possibly much more than half. But the other half is not a bubble, and is growing rapidly. What the collapse will look like depends in part on how much the productive segment grows relative to the other part before it happens.

It wasn't from "a random influencer". It was in a popular science publication, and I believe they were quoting (or perhaps paraphrasing) the person who invented the term.

Does it have a "legal definition"? I doubt it. So for regulations I think it means whatever the person enforcing the regulations wants it to mean.

Reading all these comments makes it clear that we on Slashdot have become who we used to ridicule: Science-denying zealots.

That really depends on exactly what definition you are using. I suppose you could argue that yogurt could be made at home in a normal kitchen, but cheddar cheese couldn't. And I've never actually seen anyone make sauerkraut, though people certainly used to do so.

I.e., the first published definition of "ultraprocessed" specified "things that couldn't be made in a normal kitchen". I'll agree that it's a very sloppy definition, but I haven't heard a better one.

It varies from time to time. In the 1940's it was common for folk to get tattoos, at least in the US Navy. Ozzy the iceman was covered with tattoos. (Of course, since he was killed it might indicate that he was something analogous to a criminal, but probably not.)

The real problem is that minimally processed food doesn't keep as long, and often takes more time to prepare.

Actually "ultraprocessed" is too broad a category. It includes things like cheese and yogurt. Probably also sauerkraut. But there definitely are ultraprocessed foods that should not be sold without a strong warning, and many do have deceptive advertising that appears intentionally deceptive.

Damn. You're right. That article doesn't say it, and I didn't find the one I originally read, which was about bacteria living deep in the earth where the radiation generated ionization states that they used. IIRC it was about bacteria living in a granite based low-level uranium source. And they were living a lot deeper than previously detected bacteria. (This was about 3-4 decades ago, so it's not surprising that I can't find that article. I think it was in Science News, but possibly it was in New Scientist. In any case, what I read was a magazine article. And it was rather explicit...though of course not detailed.)

Here's one reference, though not the orginal one I read.
https://academic.oup.com/ismej...

Note that this is not DIRECTLY eating radiation, but that's still its energy source, as that's where the H molecule comes from. So to simplify, saying it eats radiation is not wrong.

It's quite plausible that it eats radiation. There are bacteria that live inside rocks and eat radiation. That it would be a shield is, however, very implausible.

But if they talk, should you believe them. People say all sorts of things. You can't really trust strangers whose motives you can only guess at. Perhaps they're about to be fired, so they want to damage the company.

For that matter, if someone said a game was NOT made with AI, I wouldn't believe them. They only know part of what was being done, so even if they're intending to be honest they can't be believed.

I think he was probably correct when he asserted "AI will be a part of the way all games are made".

It's not slop everywhere else, just in many places. AIs that have been custom trained for a particular situation can often do quite well. This work particularly well in classification, but also works in several other areas.

The main criteria at the moment is "so you have an easy way to check correctness?". If you do, then AI can, when properly trained and configured, do a good job.

This is supposed to be for NON-emergency calls. Not comparable to 911.

No. It can't be properly expressive without understanding the story that it's reading. Punctuation is just not enough, it doesn't capture many different shades of meaning. E.g., an ironic statement should be read in a different tone than a factual statement, even with exactly the same punctuation. (That's one example out of MANY. Consider, e.g., the scene in "Alice in Wonderland" where she's talking about jumping off the top of the house.)

Well, I *do* want an "AI PC", but not anything currently on the market. I want one that will understand books in HTML format and read them to me in a reasonably expressive tone. I'd also like it to be able to pause and then answer questions about what was going on earlier if I missed a point.
OTOH, I'd also want it to be strictly segregated from most of what I do.