He's going full Gavin Belson, isn't he?

I thought it was more David Icke

Look, the AI models have every bit as much inherent value as the ants, I mean humans, who we billionaires were previously depending upon to do our bidding.

This narrative that it's genocidal to switch off a few million humans, but somehow acceptable for me to not switch on a bunch of planet-destroying hardware is unacceptable. It's getting in the way of me playing business.

With every new Altman utterance I get more Elizabeth Holmes vibes.

Has he started wearing black turtleneck jumpers?

Sam Altman yet again proves he produces just as much bullshit as his LLM Ponzi scheme does.

He ignores the pure and simple fact that the bullshit mills cannot create , and only mimic stuff stolen from across the internet.

Well, he is an empty vessel himself, just like his LLM.

In fairness, most humans are also mimics with nary an original thought.

Our style of consumerism is also a massive expenditure of resources which yields little-to-no lasting value, except to those who manage to take and invest a slice from each drone's consumption.

It's a Brave New World, mates.

AI consumes ADDITIONAL energy on top of that already consumed by humans... or is he suggesting a cull?

99.999% of those humans will have more intelligence than his turbo-clippy will have in all their lifetimes combined

I have suspicion managing not being eaten by the minor predators might be up there too.

Brings to mind his sister’s accusations about him.

it cannot immediately detect that Altman is totally ga-ga. For left pondians - bananas, ape·shit crazy.

All the sheep have left the top paddock; somebody forgot to pack the sandwiches; even turning out his pockets you wouldn't find ¼d let alone a ha'penny.

Or farthing. Which he is, in terms of being a long way off from humanity.

Mine’s in the pocket, rather than down the back of the couch.

"And [GPUs] do break a lot, still."

And they'll break even more, when exposed to cosmic rays.

In October 2025, Mr.Altman stated that advertising on the platform was a total last resort because it's what failing businesses/models do in a last ditch attempt to get funded.

In February 2026, OpenAI integrated and started to implement adverts in to it's answers.

You could not believe his answer about what the weather was like outside without looking out the window yourself first. So his viewpoints, however comical, are to be ignored. Or laughed at.

Man jumps shark.

Someone send Jensen some water skis...

Many people are held back due to parental, community and societal attitudes on learning, where schools are not designed to educate but instead to contain and condition it's "inmates" to produce unquestioning, compliant worker drones

Otherwise you could skip much of the dead time (which the Simpsons parodied so well with "magazine time" and Skinner's justification that he can't challenge intelligent students as the "stupider kids would be sitting there furrowing their brows, trying in vain to understand"), bin all the "rote learning" so beloved of right wing politicians and pub bores and replace it with giving kids problem solving strategies, alongside more practical skills and real life examples to get them

1) moving, 30+ hours of sitting still a week and we wonder why kids look like beach balls.....

2) working in teams Vs the obsolete "sit in silence, no talking, no discussion with others, keep your books closed" attitudes created when schools were there to provide church sponsored minimal education to young factory and mill.workers, where the church told the mill and factory owners they could have "their" workers from noon onwards. All so you know kids ACTUALLY know HOW to work in teams,.something employers have been complaining about for decades now that young folk have no idea how to work in a team

3) encourage them to ask questions, think of their own solutions

4) expose them to the real world concept that it's ok and often desirable to fail at something to foster innovation and understanding of the problem

Just in case you need to know, it's not all a conspiracy. In fact most people, most of the time, actually want to help others, or at least not actively cause them harm or disappointment. This includes many collectives/institutions.

It's always a dangerous thing to sweep people up under a label and call that label a 'bad thing', thereby inferring that the people under that label are necessarily 'bad people'.

It may suit our prejudice, ignorance and fear to do so, but it is, in fact, a 'bad thing' to do.

Added to which 20:20 hindsight is a wonderful thing, except when it is used to mock and vilify the efforts of those who were trying to make the best (or least worst) of their time and place. Remember, someone else will be mocking and vilifying you for your ignorance and stupidity in just a few years time—maybe even before you are dead and in no position to defend yourself.

Then, of course, and very sadly, there are all the genuine scumbags who only think about themselves, and are quite happy to inflict misery, hopelessness, and death on others if it happens to get them what they want. But they end up dead too, so they choose poorly for no ultimate benefit.

" the naysayers ignore the vast amount of resources humans have consumed over millennia "

How about just over the last decade? I didn't think we'd had "AI" for millennia. And 100 billion people? The current [1]world population stands at about 8 bn. This is yet another example of the unadulterated bullshit that fuels "AI investment", aka pouring other peoples' money down the drain while creaming off the top layer.

[1] https://en.wikipedia.org/wiki/World_population

> He said it was unreasonable to focus on "how much energy it takes to train an AI model, relative to how much it costs a human to do one inference query. It takes like 20 years of life and all of the food you eat during that time before you get smart," he said. "And not only that, it took the very widespread evolution of the 100 billion people that have ever lived and learned not to get eaten by predators and learned how to figure out science and whatever, to produce you."

So, he's comparing the energy cost of running ONE inference query to the energy cost of literally the entire history of life itself ?

Wow.

I... have no words.

If someone attempted this in a serious debate, I would consider it equivalent to conceding the point to the other guy.

"And not only that, it took the very widespread evolution of the 100 billion people that have ever lived and learned not to get eaten by predators and learned how to figure out science and whatever, to produce..." .... the Computer Science/Engineering which allowed for the AI model in the first place!

What a complete d*ck.

Whataboutism at its most egregious.

The solution is for AI to proactively but subtly encourage people to use fewer resources, such as eat less meat, recycle and reuse more and so on, and so pay itself off; it probably already does such thşngs to some degree...but to be fair there is some truth to the points being made, if AI is done well it does or would also pair with a saving of energy and resources in various ways. On the other hand, humans don't have a good record of taking up good opportunities to have less of an impact...

a squirrel!

"Water is totally fake," he said, given that AI datacenters often rely on liquid cooling with closed systems, rather than traditional evaporative cooling."

OK Sam, then no issue if we turn off the spigots to all the datacenters around the world. The few humans if any in the DC's, can drink bottled water and pee in the bottles after. Full recycle.

What a clown this guy is.