The AI was in an environment where it was given access to the Internet, which should not have happened in the first place. The article suggests that they think the problem was only that the AI wasn't safeguarding itself well enough, which they're going to attempt to fix by giving it new base instructions. In other words, they're still going to rely on the AI to do something it obviously can't, and are going to leave the original security holes in place.

We've seen these attempts for years and years, and now it seems the politicians are finally having some success at trying to collect everyone's personal information by claiming it's "for the children", now on the basis that "social media" is witchcraft^W addictive like heroin.

America needs to plant a moonbase first, and the problem is just getting enough petroleum there to fuel it. You know for certain that China is just going to run its moonbase on free solar energy, or what we call "Communist" energy, which is against American national security interests, but I'm sure President Trump has it well in hand and we will get our gold-plated Christian Nationalist Trump Moonbase as soon as he's finished prosecuting the vandals who tore up his beautiful American-blue painted reflecting pool, finished the ballroom, and erected the Arch d'Trumph.

Praise him and pass the collection plate! The bidding donations start at $1,000,000,000.

The aim of the game is to try to benefit U.S. billionaires. You WILL buy their inferior and overpriced products.

This sounds like some sort of weird joke, or possibly a Chuck Tingle work. I Fell In Love With A ChatBot (add bizarre obscenity).

You know, after World War II, the Soviet Union attempted to capture West Berlin by blockading it. They wound up giving way after an operation called "the Berlin Airlift", where the Allies dropped supplies into the city by parachute! Perhaps we could try that here.

Oh. We're the bad guys here? ...ummm. Shit.

Anyone? Is anyone prepared to drop humanitarian aid into Cuba?

You sound very much like you've fallen for the idea that "algorithms" are the very Devil's own addictive work.

The EC will now require books to be printed with the pages out of sequence, forcing you to "go to" the next specified page instead of reading pages in order. The existing design was too clearly addictive and might entrap young people (and some adults) into reading entire chapters at a time without considering the dire consequences.

Those kids will be prepared when 1980 rolls around again.

This is hilarious! Everybody's trying so hard to push AI on the consumer, and nobody actually wants it.

I tried doing business with them twice. Got approved, closed the deal (I thought), and got notified that the deal was canceled days later, before the vehicle was delivered. They claimed they saw some possible fraud in my paperwork, although they wouldn't explain what. The first time, I'd gone through their own bank and insurance, so I assumed one of them had screwed something up. The second time, using my own bank and insurance... got the same result. My driver's license is certainly legit, so what else could have gone wrong? They weren't saying.

I have no idea what kind of scam they're running but, they don't seem to be in business to sell cars.

Aside from being gratuitously ugly, perhaps they should have gone with the puck. I mean, they only last for an alleged four hours at a time. So, you'd need two pair if you wanted to wear them all day?

What do these weird toys do that would justify that kind of expense?

Apparently, you've never read Ayn Rand. There are worse authors but, she had such a distinct way of demonstrating that 20 pages of leaden prose can be an unrelenting horror. She was, at least, understandable, though you knew that her 20 pages could be summed up completely and concisely in a haiku.

Citation needed. They don't seem to be able to see the past or the present, let alone the future.

Are you in marketing? You're not actually familiar with how LLMs have already reached a dead-end that can't be fixed by adding more of the same.

Oh, now, don't be so pessimistic. I'm sure that, in time, AI will also be able to solve the Trolley Problem. Perhaps we'll be able to get rid of the entire field of Philosophy if we can just teach AI how to get drunk!