Sounds like I stepped on a nerve. Here, have some links.

Efforts at designing an educational curriculum that organizes subjects based on neurological development is not even remotely "soft bigotry." It's just optimizing teaching to match peak learning capacities. I don't know where you get this "single study" idea from, but I will point out that the phrase "critical thinking skills" is a very sloppily-defined high-level rollup of many mental abilities combined. Teenagers clearly have the beginnings of some of them. But plenty of evidence from unrelated studies shows significant neurological differences in adulthood that are poignant to this very topic.

Boring kids and making them reject reading is not helping them. Engaging kids and getting them excited about reading is. This is supported not only by science but by the lived experience that most of us had reading those classics and being bored out of our minds by them when we were in school, only to re-discover them and be totally moved by them later, in adulthood.
 

I read that the theory is called "priming" and it goes like this:

The raw neurological capacity for critical thinking skills don't develop in the human brain until 20 years of age (and don't really complete development until around 25). So, in an effort at making that happen sooner and better, we prevail upon young teenagers to read famous "classics" that were written for adults (books like The Grapes of Wrath or The Great Gatsby or etc.). Further, the teacher leads class discussions about the book by asking probing questions and setting the historical context and giving writing assignments about the story's events. This is supposed to stimulate early development of the neural connections associated with critical thinking.

And it doesn't work. AT ALL. It just leaves the kids insanely bored. They can understand the overt plot (which is boring) and the historical setting when told about it (but they can't relate to it because it is nothing like what they have experienced so far in their own lives), but the real substance of what these works contain is just lost on them. Worse yet, it makes them DISLIKE reading! These negative associations motivate them to look for other forms of entertainment so they never really exercise those skills.

If we instead gave them reading material that was more relatable AND more entertaining by virtue of being written specifically FOR their age group, they will actually enjoy reading it. Even if they learn nothing else from it, the sheer fact that they are engaged while reading it means that they will be pushing the boundaries of their attention span, improving their vocabulary, exercise reading comprehension skills, and etc. And best of all many of them will seek more books to read for pleasure after having a positive experience like that, so they will be master readers by the time they hit college and are actually neurologically ready to read stories about the great depression or class struggles, and actually understand their significance.

This is just one element contributing to the problem. Others have pointed out many other elements that also contribute.

Teaching your kids that it is not acceptable will not stop them from doing it.

Young kids only understand consequences. Many of them get good at acting guilty and crying their way out of punishment whenever they get caught. They learn that the magic words "I understand why it was wrong" reduces potential punishments by tenfold. And then they turn around and do it again the moment they think they won't get caught (and have some reason to do whatever it is).

It's still good to teach ones' kids about unacceptable behaviors. Eventually they start to grow a moral conscience and actually internalize the guidance. But that can't be relied-on as a method of stopping kids from doing things, because kids break rules. It's part of being a kid.

Are you talking about the scientific method? Asking other humans to do the same test to see if they get the same answer?

AI can do that too. They just need an api for doing the test (whatever it is), and they can easily have multiple separate instances of a model use the api and do the test, and then compare notes.

Most humans are not well-positioned to be running large hadron colliders, however. Mostly all they can do is read up on the information available online. And....AI can do that too!

Humans don't have some magical ability to just intuit truth from falsehood. And neither does AI.

So, my point stands.

There IS a difference in how humans are much quicker to state that they don't know something, whereas AI models tend to just make stuff up. This doesn't seem to be the exact same thing as being able to determine truth from falsehood, however. Incidentally, humans also sometimes make stuff up when they don't know the answer (key evidence: widespread belief in supernatural beings that want 10% of your income).

I already know everything worth knowing about the Dunning-Kruger effect.

Human employees cost a lot. AI is the holy grail of eliminating labor costs. As with any investment, it will cost a lot more in the short run than having human employees would, but once the next threshold is surpassed, the savings and profits will more than make up for it.

In theory, anyway.

The slightly less cynical take is that AI can empower humans to achieve more than they otherwise could. Rather than eliminating jobs, we get a lot more productivity from the same number of employees, thus producing everything that everybody wants even faster and in greater abundance.

Either way, it's clearly valuable to many people for many reasons, so it will naturally garner a lot of investment money.

Incidentally, humans aren't very good at telling facts from lies either.

I don't think this makes the next level of AI any easier to achieve, but I don't think "the ability to tell facts from lies" can be postulated as a defining quality of a human, nor something that differentiates humans from AI.

I don't think Meta can have it both ways like this. Either they respect their employees, or they don't.

I am just glad I don't work for Meta. I feel sad for you all. I hope you can find something better.

The evil for which we hate Microsoft is also the reason that we all know the name. They won't ever stop being evil because that's exactly how they maintain their wealth and power.

Well I don't know, I am not a lawyer. I know that there ARE laws that penalize malicious code, like the Computer Fraud and Abuse act, and that there is a legal concept of "protestware" that draws a distinction between a bug that causes data loss and intentional sabotage. And there are also license terms that disavow responsibility for any damages that may or may not hold up if intentional sabotage is established.

But all of this is outside of my domain. I was speaking generally, that there are laws against malicious software writing/distribution in general.

Based on what I read of this story, it sure sounds like intentional sabotage to me. An instruction to an AI that orders it to destroy data sounds like a "weaponized prompt" to me. It's one thing to say "this software might have bugs or not even work at all, and those bugs might cause harm, so that's your risk to take." It is quite another to deliberately code malicious prompts as a trap with the intent of harming those who fall into it. I don't think someone can do that and say "well my hands are clean because I said you use this software at your own risk." Even if the law is gray in this area because AI is new and emerging, the social principle here is pretty clear.

The original star wars movies had many elements that drew in audiences at the time, including a plot about a mystical force that was guiding a new hero on a path to save the galaxy from overwhelmingly oppressive tyranny. The events were significant and the family-tie shockers injected some drama and so they were good.

But "Star Wars: The Last Flop" lost the thread. Instead of a plot that was even more epic and had even more galactic significance, it just doubled-down on the family drama and kind of lumbered around, getting us nowhere new. There was plenty more to dislike in terms of how they ruined character arks and pushed a political agenda that did not sit well with much of the audience.

Ever since then, the franchise has been sliding downhill. I read summaries of the other movies and shows and they all sounded equivalently vapid. I think I am not alone in this opinion.

You are familiar with the idea that the tool should match the job? And in particular, that some tools are very poor fits for some jobs, but that doesn't make them bad tools (since they are intended for a different job)?

MongoDB exists solely to remind us that some tools are not right for any job.

As it turns out, there are laws against deploying malicious code, too.

And for similar reasons.

It is true that people should vet the code they use. But this truth does not give coders license to code malicious Trojans into their offerings.

How would you feel if your grocery store deliberately put poison into the food and then after you got sick said that it is your responsibility to test all the food you buy?

"I didn't kill him, the bullet did."