I'm trying to think of how you monetize eyeball scans and the first thing that comes to mind (well, after being able to break biometric security) is training an AI to generate fake but passable eyeballs to undercut the use of iris scans as an anti-bot tool.

Like, it's such a specific piece of data and it's not going to provide much utility for advertising targeting or profiling or whatever else, so I'm not really sure where the money is on this one. Not sure if I'm missing something or if this is another case where I'm trying to assume these people have the barest idea what they're doing.

See, I feel like AI might have the actual solution to this problem. We can overcome the economic issues with setting up SMR infrastructure the same way AI has powered through all their economic problems: setting VC money in fire and trust that the smokescreen will hold out for another funding round.

Once the reactors exist, I'm assuming that their operation can be relatively cheap for whoever ends up owning the actual plants once the AI bubble pops and the datacenters around them are shut down or repurposed.

Man, who could have predicted that entering a close business relationship with Microsoft as a software company was a potentially bad idea.

At least the product is AI so Microsoft's incorporation is largely guaranteed to be a massive loss.

Honestly I feel like trying to find the root of his bullshit and how specifically it's so radically off-base is actually kind of an important part of the project Otherwise the follower crowd will just claim that he's not being given a fair shake, when he was obviously talking about (insert a brief and misleading explanation of the circle thing that actually does try to make it make sense).

That's what the whole narrative is built on, after al I'll.

Common mistake, but David is from the UK. He was actually just looking online for ways to improve wheelchair accessibility in a multistory house. On a lift-shopping website.

Did Sam anticipate the easily foreseeable avalanche of AI slop, decide that proof of humanity was a worthwhile investment, and only then notice that all the suggested search completions for "proof of" were crypto?

But users like engagement! Which means seeing that the person wrote a thing and not reading the thing and making a human connection with someone who is, y'know, engaged in conversation.

I know chatbots don't track meaning, but I'm pretty sure words still mean things.

I can also hear Timnit Gebru wailing in despair all the way from my house, because this is exactly the kind of AI-reinforced bias that she was talking about while everyone was arguing about whether to listen to the doom cult. These are not theoretical problems with the technology; they are fully extant and easily demonstrable. The more these systems are integrated into processes that effect actual people the more often we'll see this kind of thing happen.

Less well known, but perhaps more interesting given that the phrase blue blood has already been used.

That part almost makes sense. Like, from the Bank's perspective your deposit is a liability, not an asset. But the mortgage on your house that they very much can give you is an asset for them and a liability for you. Two-column accounting works like that; everything in my asset column is going to be in someone's liability column.

But none of the other nonsense follows from that at all. It's the few sentences bordering on lucidity to make you forget how insane all the promises from earlier were.

It's true. Also notably the scam isn't on the car dealerships or the family court or, it's on the guy who buys a course in sovcitery, or the books, or the document templates, or whatever else they're selling.

I mean, of course they didn't carve "this" - they didn't speak English, so "this" would have just been a random assortment of squiggles!

I wouldn't say there's no mysticism in the singularity, at least not in the sense you're implying here. While it uses secular and scientific aesthetics rather than being overtly religious the roadmap it presents relies on specific assumptions about the nature of consciousness, intelligence, and identity that may sound plausible but aren't really more rational than having an immortal soul that gets judged at the end of days.

And it doesn't help that when confronted by any kind of questioning of how plausible any of this is there's a tendency to assume a sufficiently powerful AI can solve the problem and assume that's the end of it. It's not less of a Deus ex Machina if you call it an AI instead of a God to focus on the Machina instead of the Deus.

Except their magic money machine is fake, but their need to buy food and housing and whatever else is very real. In order to sustain themselves they're either also grinding away at a 9-5 in addition to trying to make the magic happen or else they're straight-up grifting other sovcits and aspiring sovcits into giving them money.

There's a reason why the C in OPCA stands for "commercial".

No no, they're not breaking the law they're using their secret law magic to let them follow the Superlaws that apply to the coolest and sexiest people rather than the plebian laws that apply to people like the cops or the courts or their ex-wives.

I mean, if you want to trace it back that far you're talking about the bronze age Canaanites, and arguably through them the Phonecians and Carthaginians would be better successors than the Israelites. If that's the case then the land should either be under Tunisian administration (for maximum cultural continuity after the fall of Carthage during the Punic wars) or either Turkey or Italy for political continuity through right of successive conquests. Or maybe we want to go with one of the theories about just how far afield the Phonecian diaspora went and consider the state of Utah their most direct modern descendent.

The ancient history of the levant is fascinating, but in the context of the current conflict over the nation of Israel it is relevant only in that it is a major part of Israeli nationalist mythmaking. If you want to actually trace who lived in the region thousands of years ago and where they ended up to determine who the "rightful" government should be you're going to end up in some really goofy places unless you already know what answer you want to get and are seeking justification rather than illumination. Hell, nationalism as a concept only dates back a few hundred years, and before then you're arguing about the competing claims of kings and empires through descent, conquest, oaths of fealty, and whatever else. Who ruled/owned/managed your land and who you were as a people were far less connected, and using that as a justification for modern government actions honestly sounds like they're out of actual justifications. It's just Putin ranting to Tucker Carlson about medieval maps for two hours while Russian troops shell Ukrainian cities.

Nah, his ego is big enough that he's probably some version of the an-cap, with a personal political based in "I could give everyone this utopian world if the government and the 'woke mob' and the labor organizers would just get out of my way."

The first image in that second link is perhaps the most incoherent political cartoon I've ever seen. Why is Uncle Sam as played by Angry Jeff Bridges wearing the Chinese flag as a cape??

Thanks to the power of Technology(tm) we can have an LLM generate spam, an automailer send it out to millions, where an automated spam filter can identify them and hide them in a separate inbox to be automatically deleted in a couple of days. Of course the technology isn't perfect and sometimes someone sees one of these ads and, I assume, spends money on a product. But I have faith that these problems are solvable and we'll be able to totally automate email spam to no longer interact with human beings at any point. Once that's done we can apply the same methodologies to weekly internal memos, daily team meetings, and even unsolicited dick pics. Imagine never needing to take or see pictures of a stranger's junk ever again while still massively scaling up the number of unsolicited dick pics in flight at any given time.

I guess the other relevant example in film would be Wargames. At least nobody is recommending hooking ChatGPT up to the nuclear launch system.

I honestly hope you're right about the coming AI winter. I was watching a report from a major South Korean arms show yesterday and one of the themes that the defense industry appears to be taking from the war in Ukraine is that more independent and autonomous weapon systems are going to be useful tools to counteract jamming and to reduce manpower requirements. Nobody appears to be putting OpenAI into the kill chain, and I think on balance autonomous systems are a more ethical way to address the problem than other options. I can't believe we're seeing cluster bombs and land mines making such a comeback. All the same, a strong push against AI more generally would help make sure we don't end up combining the IFF of a land mine with the trigger discipline of the CIA's predator drones.