WhO cOuLd HaVe PoSsIbLy PrEdIcTeD tHiS?

I think the association of philosophy with abstract imagery is because the only time that the vast majority of people of the Baby Boom generation ever got "philosophical" was when they were high.

None of this shit is inevitable. The people saying it's inevitable want it to be inevitable, so they're trying to make it inevitable by claiming it's inevitable at every opportunity, so everyone will just resign themselves to its inevitability and just start using it.

Further, AI aside, in the vast sweep of history, technology has not been some unalloyed good. Everything's a trade-off. Plumbing and electricity and automobiles and airplanes and semiconductors have got their upsides and downsides. Almost everybody likes medicine, sure, but fewer and fewer people can afford it. The major quantitative benefit of technology has been to let us support an ever larger population. But then the population always grows to take up the extra capacity, and then you're back in the same boat but it's more crowded and leakier.

I mean, they are losers, in the sense that the oligarchs, of which this guy is one, have announced repeatedly that most of them are not going to have jobs because of this new technology that he finds exciting because it allows people like him to finally be capitalists without all those pesky workers and their pesky mouths. That was a significant loss for these college kids. Hence, losers. Shut up, losers. Quit losing all the time, losers! End of the world, environmental devastation, billionaire AI bros competing to build the biggest apocalypse bunker, bitch, bitch, bitch.

Because this guy says there's going to be a productivity tsunami! You, too, in fact everyone, can be a software oligarch! All you have to do is strap yourself to a plagiarism machine long enough to train it to replace you. Just relax into it and watch your skills degrade in real time. Your craft, your mastery, your pride and worth and dignity as a human being? Just shut the fuck up and ask Copilot. Watch it poo out some code you no longer understand how to review, you dummy. Slap a Midjourney-generated logo on it and throw it on the pile and sell it to... uh... well, nobody actually has any money. But surely that problem will solve itself somehow. If you just stop being such losers!

Surely you've noticed how great software has been getting lately! Certainly you haven't noticed a bunch of dev teams bragging about not having written a line of code in four months, while their website simultaneously drops from five nines to nine fives of availability.

Or we could use political pressure to ban AI.

"Everyone must stop putting lead in gasoline, in paint. Stop using gas appliances in homes. It's bad for you." Sure. I can accept these policies. In the grand history of humanity, these were relatively recent changes, and we have better alternatives now.

"Everyone must stop drinking alcohol in any amount. Stop burning wood in any context. It's bad for you." I don't know. These practices predate history. It's not rational, maybe, but I am extremely hesitant to support banning these practices, even though I don't do either of them myself very often. I feel like these are part of what being human is. We've evolved around them.

Sometimes dogs bite. Maybe you can prevent your armpits from sweating with botox. But there are some things you put up with, because you wouldn't be quite human without them. A perfectly rational animal would just crawl inside a mechanical womb and never come out. But not a human.

It's well-known that laying workers off decreases short-term expenses and therefore improves the apparent short-term profitability of a corporation, which in turn makes the executive team appear to be performing better on the metrics on which their continued board support and compensation are based. So it's strange how nobody ever gets laid off for those reasons.

When the bubble collapses, are they going to bill everybody?

'The company says it received "digital confirmation" that the information was destroyed and that affected schools and students would not be extorted. The BBC reports.'

For a company that makes education software, they sure must think their customers and users are pretty stupid.

Rest assured, if I ever attempt to activate an AI feature, I am currently having, or have already had, a severe stroke.

According to the GIF of the proposed interface in the linked article, you can shake the mouse cursor back and forth to activate a mode where clicking anything "asks Gemini about it". If ever there was a false mapping between gesture and function, this is it.

Back in the old days, when Nvidia was just a company that built graphics cards for carding graphics with, I always wanted them to admit to themselves that they were a hardware company, who sold hardware, which is where their money comes from, so they wouldn't feel the need to sue people trying to write drivers for their cards. So this reinforcement of their self-identification as a software company isn't particularly welcome to me. But c'est la something or other, as the French say.

"The tunnel segments are 217 meters (two football fields) long, weigh more than 73,000 metric tons (slightly more than more than 73,000 REAL tons), and have to be placed within a tolerance of 3 mm (0 inches)."

Fixed. Thank you for bringing this oversight to our attention. Note, however, that this still lacks an analogy to liken this tunnel-for-cars to an actual car somehow.

"The tunnel segments are 217 meters long, weigh more than 73,000 metric tons, and have to be placed within a tolerance of 3 mm."

I simply can't imagine why it wouldn't be orders of magnitude cheaper and faster to just use short segments that don't need to be aligned so precisely.

I used to think that maybe I, a simple country ignoramus, just wasn't equipped to understand the Wonders of the Modern Age. But the more glimpses I snatch of the wiring and plumbing of the Wonders, the more I think somebody in administration WANTed this to be unnecessarily expensive because their butt-buddy got the contract.

As I understand it, LLMs remembering user directives only selectively has been one of the major complaints that users have with LLMs.