If I were in charge of the defence department of a government, there’s no way I’d ever allow closed source software to even touch machines with sensitive data. Or even pre-compiled software. Because the problem is not what you see, like personnel; it’s what you don’t, like bribery and backdoors.

Let alone use “cloud” computing. Come on… it’s someone else’s computer.

There are some explanations here, but I can’t follow any of them.

First you start with a lace pattern: make red and white strands of glass, and twist them together. Like this, but imagine it’s hot glass (not too hot to be liquid, but hot enough to be malleable):

Then do the same with blue and white. Finally, lay them side by side and heat it up until they fuse together. You’ll get a “sheet” of checkerboard glass; now all you need to do is to gently heat it up just enough to shape it into a plate.

deleted by creator

The blue in the plate is likely Egyptian blue, the first synthetic pigment humankind produced. Made by baking together sand, natron, and grated copper.

The red is likely cinnabar. Potentially mixed with rust. Scrap that, likely gold, it turns red in glass. Cinnabar wouldn’t survive the high temperature, and rust would give it a green tint.

Not sure on the white. It could be lead white or simply chalk. I’m betting the former because this looks expensive.

If mods aren’t games, then gravity doesn’t work on Fridays. Dumb arbitrary restrictions being pulled out of nowhere.

Most likely:

The overall placement of the stars fit a mirrored version of the carving rather well. They even made Antares (yellow square) larger=shinier than the others.

I don’t like pizza cutters because they never cut the borders properly, specially if I prepared the pizza in a certain rectangular dish I use often. So instead I use a knife with a curved blade - likely the same I used to prepare the ingredients to add to the pizza.

For delivery pizze this isn’t necessary, they’re already sliced.

I’ll probably watch it but it lost the shine from s1.

It is not impressive, but it makes the rest of the output even worse. You’re expected to treat the bot’s output as human language, but it doesn’t make sense like language would: it identifies the soy sauce, it should be able to identify the bowl is empty, no change happened, and yet it’s still babbling that the guy “already combined the base ingredients”.

[Guy] Spank me, daddy!
[Zuckerberg] Current location of your male parent required as further info.
[Guy] Ah, come on, just hit me Zucky~
[Zuckerberg punches the guy on the teeth]
[Guy] BLAME THE WIFI! BLAME THE WIFI!

…sorry I couldn’t resist.

My take on this matter is here and it includes a transcription. I’ll only address the HN comments here (I can’t be arsed to check the Reddit thread.). You know, for my five seconds of rambling.

[HN] That wasn’t prerecorded, but it was rigged. They probably practiced a few times and it confused the AI. Still it’s no excuse. They’ve dropped Apollo-program level money on this and it’s still dumb as a rock. // I’m endless amazed that Meta has a ~2T market cap, yet they can’t build products.

In the hypothesis it’s rigged, it’s even worse: there’s additional malice, but the rigging sucks so much the incompetence is still there.

Also, market cap has nothing to do with overall usefulness to society, so I’m not surprised some dead weight company (YES) has a large MC.

> confused the AI. // I will die on this hill. It isn’t AI. You can’t confuse it.

The comment is just an “ackshyually” but it happens to be correct.

The Kotaku article on this had a really nice final zinger[0]: // > Oh, and here’s Jack Mancuso making a Korean-inspired steak sauce in 2023.

Completely off-topic: I think sauce on steak is blasphemy. Not even chimichurri (less exotic for me than barbecue sauce) - it goes nice on some sausages, but not on steak. That’s a personal take though, it isn’t like anyone needs to agree with it.

[1] Credit where it’s due: doing live demos is hard. Yesterday didn’t feel staged—it looked like the classic “last-minute tweak, unexpected break.” Most builders have been there. I certainly have (I once spent 6 hours at a hackathon and broke the Flask server keying in a last minute change on the steps of the stage before going on).

[2] Live demos are especially hard when you’re selling snake oil.

[3] Live demos being hard isn’t an excuse for cheating.

[4] Despite the Reddit post’s title, I don’t think there’s any reason to believe the AI was a recording or otherwise cheated. (Why would they record two slightly different voice lines for adding the pear?) It just really thought he’d combined the base ingredients.

1: No, no credit because it is not due. “X is hard” is not the same as “you should get credits when you fail X”.

2 and 3: spot on.

4: I wouldn’t be surprised if it was staged, I do not think it’s above Merda Meta to do so, even if I believe it’s simply incompetence due to Hanlon’s Razor.

This does not deter me from possibly buying one. The concept is pretty cool and appealing to those who want a distraction free lifestyle. Even if there’s a screen in front of you at all times, at least you won’t need to hold something in your hands to be able to operate it. That alone is a significant win.

Your money, your problem: you buy what you want.

But my opinion, my problem: I think you’re a sucker, because the quote below is likely spot on:

[note: clipped comment] It’s an ad network with an attached optional pair of glasses. // It’s the platform Zuck always wanted to own but never had the vision beyond ‘it’s an ad platform with some consumer stuff in it’.

Bingo.

I love how they randomly blame the WiFi network, like anyone is going to buy it.

It’s almost certainly a joke. Everyone knows that the demo failed.

Even in the hypothesis it was a joke, the fact they don’t highlight “hey, it’s just a joke, the model failed” right off the bat shows they’re eager to mislead you. And yes, you can use jokes to mislead someone.

[1] So much negativity. // I’m just excited that our industry is lead [HAHAHA] by optimists [HAHAHAHA] and our culture enables our corporations to invest huge sums into taking us forward technologically. [HAHAHAHAHAHAHAHA] // Meta could have just done a stock buyback but instead they made a computer that can talk, see, solve problems and paint virtual things into the real world in front of your eyes! [PFFFFFTHAHAHAHAHAHAHAHAHA] // I commend them on attempting a live demo.

Yes, the mocking, gleeful negativity really does make me concerned that this place is becoming Reddit. The fact that the highest upvoted post on this thread is just a link to Reddit isn’t doing much to help me feel better. And I’ve been here for at least a decade, so I don’t think this is the noob illusion.

1: Laughs added by me. Oh wait, he’s serious, let me laugh even harder.

2: What makes “Hacker” News already extremely similar to Reddit is not the “negativity” you’re babbling about. It’s the fact most comments are skibidi in the same ways:

• shallow stuff even a 5yo could come up with. No, wait, “even AI could write this”.
• they’re irrational = fallacious = disgusting. On the same level as “I dun unrrurstand, u think 50 is not 100? Than u think 50 is 0? dats dumb lol lmao”. They get caught on the words instead of the concepts being conveyed by those words, they vomit assumptions and re-eat their own vomit, false dichotomy and genetic fallacy are rampant there.
• “Ackshyually” here, “ackshyually” there… just admit “I’m too much of a moron to understand what would be a meaningful contribution, so I’m vomiting irrelevancies and pretending they’re meaningful”.
• sealioning everywhere. Holy fuck.

There’s no hacker culture in “Hacker” News. It is not made for the curious-minded, exploring some intellectually stimulating topic. It’s made for braindead morons who really want to believe they’re insightful, and since [sarcasm]filthy reality bends to their pristine wishes then they must be insightful, right? [/sarcasm] The only things missing from Reddit are 1) scope (Reddit has more topics to talk about), and 2) “WOW THANKS FOR LE GOLD TO LE EPIC KNEE KIND STRANGER! THE NARWAL BACONS!”

inb4 me being a grumpy fuck doesn’t what I said above true or false.

One important thing to note: demo didn’t fail! (Or, at least not in the way people usually think of)

Yeah. The demo didn’t fail. Suuuuure. And the Moon is made of green cheese.

Senku was petrified for seven years.

The wiki says Suika was 12 when everyone was petrified, 18 afterwards, so it took her six years. Based on vegetation growth and my eyeballed estimates it would be four instead; the tower looks ~8m tall and lianas grow ~3m/year, so it would take them at least three years to reach the top.

We got it. We’re roughly at chapter 196.

Transcript:

• [guy] Hey, Meta, start LiveAI.
• [two centuries later…]
• [robot] Starting LiveAI. I love the setup you have here with soy sauce and other ingredients. How can I help?
• [guy] Hey, can you help me make a Korean-inspired steak sauce for my steak sandwich here?
• [robot] You can make a Korean-inspired steak sauce using soy sauce, sesame oil…
• [guy, interrupting bot] What do I do first?
• [three centuries later…]
• [guy, repeating] What do I do first?
• [robot] You’ve already combined the base ingredients, so now grate a pear to add to the sauce.
• [guy] What do I do first?
• [audience laughs]
• [robot] You’ve already combined the base ingredients, so now grate the pear [audience laughs] and gently combine it with the base sauce.
• [guy] Alright, I think the Wi-Fi might be messed up. Sorry, back to you, Mark!
• [robot LARPing as a guy] [Mark Zuckerberg] It’s all good. Youknowwhat? It’s all good. The irony of the whole thing is that you spend years making technology and then the Wi-Fi at the e[nd of the] day kinda catches you.

My comments:

1. Wi-Fi my arse. This is blatantly bull fucking shit. The model answered the situation wrong; it is able to parse individual items in the footage (note how it praises the “setup” at the start), but it babbles about the guy combining the base ingredients even if not the case.
2. Bot feels like a slowpoke. Seriously, it takes ages to answer the guy.
3. Anyone with a functional brain knows those models don’t understand shit. However, answering “what do I do first?” with the assumption a person already did some steps is dumb even for those models.
4. People don’t repeat questions to get the same answer. Is the “context” window of the bot that small?

I liked it. It was obvious for the viewers, but Suika was still a child, and it’s how children think - they want easy and fast solutions. It also shows well that with science you don’t get ir right the first time, you need to be a bit stubborn.

Also, all the screenshots are done by hand, so as long as you indicate which one you want clearly enough for a human to understand, it should be ok.

Good to know! Thank you!

The core argument of the text isn’t even arms race, like yours. It’s basically “if you can’t get it 100% accurate then it’s pointless lol lmao”. It’s simply a nirvana fallacy; on the same level of idiocy as saying “unless you can live forever might as well die as a baby”.

With that out of the way, addressing your argument separately: the system doesn’t need to be 100% accurate, or perfectly future-proof, to be still useful. It’s fine if you get some false positives and negatives, or if you need to improve it further to account for newer models evading detection.

Accuracy requirements depend a lot on the purpose. For example:

• you’re using a system to detect AI “writers” to automatically permaban them - then you need damn high accuracy. Probably 99.9% or perhaps even higher.
• you’re using a system to detect AI “writers”, and then manually reviewing their submissions before banning them - then the accuracy can be lower, like 90%.
• you aren’t banning anyone, just trialling what you will / won’t read - then 75% accuracy is probably enough.

I’m also unsure if it’s as simple as using the detection tool to “train” the generative tool. Often I notice LLMs spouting nonsense the same model is able to call out afterwards as nonsense; this hints that generating content with certain attributes is more complex than detecting if some content lacks them.

[Posting this in a separated comment to not confuse rikka]

This scene was bloody amazing. A damn great adaptation of the manga:

The anime expanded the time Suika is alone in the world; it was just three chapters (194~196), but we got a full episode out of it.

And I’m glad it did. It doesn’t change the plot at all, but it gives Suika’s time alone a well-deserved depth.

Sure, she woke up all alone, just like Senku did seven years earlier. But unlike Senku she was still a child, and the episode showed well how lonely and vulnerable she felt. (Specially the part where she hugs Kohaku’s statue.) And Suika was never shown to be a talented scientist or anything similar; she didn’t even get modern education. And yet she was able to make the revival fluid. It plays really well with the theme of the anime, on science being not quite the result of a few talented individuals, but of knowledge accumulated over time: previous knowledge (Senku notes), failures (the rain over the nitrate crystals), and eventually success.

By far one of the best episodes I watched this season.

@[email protected]

[OP, sorry for the harsh words. They’re directed at the text and not towards you.]

To be blunt this “essay” is a pile of shit. It’s so bad, but so bad, that I gave up dissecting it. Instead I’ll list the idiocies = fallacies = disingenuous arguments it’s built upon:

• Nirvana idiocy = fallacy: “unless its perfect than its useless lol lmao”.
• Begging the question: being trained on [ipsis ungulis] “the entire corpus of human output” with enough money to throw at it won’t “magically” make AI output indistinguishable from human generated content.
• Straw man: if the author is going to distort the GPTZero FAQ, to double down on the nirvana idiocy, they should at least clip the quote further, to not make it so obvious. There’s a bloody reason the FAQ is focusing on punishment.

Note nirvana fallacy is so prevalent, but so prevalent, that once you try to remove it the text puffs into nothing. The whole text is built upon it. (I’m glad people developing anti-spam systems don’t take the same idiocy seriously, otherwise our mailboxes would be even worse than they already are.)