Oh, okay, silly me. Don’t believe I’ve seen it.

Eh, I don’t fault institutional kitchens for the rectangles. When you have to feed a ton of people, maximizing oven space is important.

He gets shit for drone strikes (which he increased but didn’t start) that had civilian casualties, but no mentions that everyone prior just used larger dumb bombs that had a lot more civilian casualties. Everyone just says “Look at how many drone strikes under Obama!” Here’s the estimated number of middle eastern civilian deaths since 2001:

From this article

That looks like the pizza my HS cafeteria would serve, except it doesn’t appear quite as greasy.

No clue. Not making any claims other than that there’s no known expiration for water that’s not subject to excess heat.

God I miss them. He wasn’t a perfect president, but it seemed like he really tried to do the right thing.

Right, and doing that can make it go above levels even before the best buy date. But bottled water that isn’t allowed to get really hot doesn’t have a known expiration.

It’s a pretty low risk unless the bottles get really hot.

There are different types of dates in the US. Few things have expiration dates, which means it can be dangerous (or, for medication, ineffective) after that date. Most things have “best before” dates, which means the company has tested the product that far from its production and found it still met the quality standard.

The problem is that the FDA requires that testing and that every product have such a date. People have mentioned salt, which is inert, having a date, and that’s probably the most ridiculous example, but there are lots of things that have super long shelf lives beyond the best buy dates. Honey, soy sauce, bottled water, and vinegar being examples that come to mind.

But if you read the article, is not what we assumed. The boxes were being shipped from the German embassy in Japan to Argentina, and were marked as personal effects. The Argentinians did a spot check and found the propaganda, so they confiscated it because they were worried it could impact their neutrality. It went to the courts, up to their supreme court, though no one knows what action if any the court took. Obviously the Nazis didn’t get the material back.

Argentina being a safe haven for Nazis didn’t happen until after the war, if I understand things correctly.

Not necessarily. My wife and I broke up (separated) but we didn’t divorce.

Oh, for sure. It’s not an area that I’m an expert in, but your conversation got me curious because I had heard both things, so I read a few articles. One of the interesting things is that the bit about Hitler’s envoy wasn’t broadly known until 2020 when the Vatican released a whole mess of documents that had been kept hidden previously. So there’s maybe more reason for some people to have one impression based on what they grew up hearing, and others to have a different one based on more recent info, but even with the new info it seems nuanced.

It’s easy to fault anyone who didn’t take a clear stand against Hitler from this vantage point, but it must have been hard to be a world leader facing the possibility that Hitler would be successful and you’d have to deal with his empire. Pius XII supported the allies and it’s obvious he was against Hitler, but he was reluctant to be overtly vocal about it, and he even entertained the envoy, which maybe he saw as hedging his bets (we can’t know exactly what he was thinking). It’s for sure stained his legacy.

It’s less black and white than either of you guys are implying. You’re taking about Pope Pius XII, who was Pope from 1939 to 1958. He’s credited with saving hundreds of thousands of Jews through various means, including support to the resistance, but also by showing some had been converted/baptized, which in itself was controversial. But he also feared the Nazis were going to win the war and that the church would have to exist in a Nazi world. He never clearly spoke out against the holocaust, and though a lot of the things he said, including his sermons, can be taken as condemnation of Jewish persecution, the language was pretty subtle. Also, Hitler had an envoy secretly meeting with Pius from pretty early on, which many say was bad in itself. Still, he also helped the allies with intelligence, to the point where Hitler accused him of espionage and had plans to kidnap him.

I.e., they’re not programmed to listen to our feedback in a meaningful, educational way

Right, because “listen” and “educational” don’t apply to a software application like this. It has a model based on processing a truly huge amount of text. Your lone correctional prompt might tweak that model, but only slightly.

And sure, as a tool, LLMs can be very useful. I managed a software engineering organization for an aerospace company for a lot of years, and I made a number of constraints about how the LLM could be used (the company had one inside the firewall, so there weren’t IP issues), but I for sure encouraged it to be used. Essentially I was concerned about our software engineers using it in any way where they counted on it to be correct, because it often wouldn’t be. But it was great for things like suggesting test cases to test a piece of code.

That’s right, you have it exactly. When the prompt is that the prior output is wrong, the program is supposed to apologize and reprocess with a different output, but it uses the same process.

We’d be better off talking about AI if no one used words like intelligence, cognition, think, understand, know, learn, etc. They don’t do any of those things.

People really misunderstand what LLMs (Large Language Models) are. That last word is key: they’re models. They take in reams of text from all across the web and make a model of what a conversation looks like (or what code looks like, etc.). When you ask it a question, it gives you a response that looks right based on what it took in.

Looking at how they do with math questions makes it click for a lot of people. You can ask an LLM for a mathematical proof, and it will give you one. If the equation you asked it about is commonly found online, it might be right because its database/model has that exact thing multiple times, so it can just regurgitate it. But if not, it’s going to give you a proof that looks like the right kind of thing, but it’s very unlikely to be correct. It doesn’t understand math - it doesn’t understand anything - it just uses it’s model to give you something that looks like the right kind of response.

If you take the above paragraph and replace the math stuff with therapy stuff, it’s exactly the same (except therapy is less exacting than math, so it’s less clear that the answers are wrong).

Oh and since they don’t actually understand anything (they’re just software), they don’t know if something is a joke unless it’s labeled as one. So when a redditor made a joke about using glue in pizza sauce to help it stick to the pizza, and that comment got a giant amount of upvotes, the LLMs took that to mean that’s a useful thing to incorporate into responses about making pizza, which is why that viral response happened.

Normally: awfully darned funny. When sleep deprived: hardly funny at all.

I had a Sound Blaster 16 in my computer from about that year. Had totally forgotten that sound cards were almost as prevalent as video cards.