trinket

@crossoverpotential.bsky.social

21 | it/any | the cat doesn't shut up

Many Mario games use the same Mario sprite introduced in Super Mario World. Despite this, in each of them, Bowser is drawn differently, so that the same Mario is pitted against 10 different Bowsers that are all meant to stylistically match him.

Super Mario World
Yoshi's Cookie (SNES)
Mario's Time Machine (NES)
Mario is Missing (NES) Bowser has two designs (intro and boss battle)
Mario's Time Machine (NES) two early Bowsers found in game's files
Mario is Missing (SNES)
Mario's Time Machine (SNES)
Super Mario Maker
A single Mario design vs
ten different Bowser designs all intended to stylistically match it
Source: Tumblr user "geno2925", geno2925.tumblr.com/post/173657736454

Bring back Heinlein from the dead to explain to him that the president is using an AI to decide if we should go to war, there's a rocket-building billionaire in control of the AI, and watching his head explode when I tell him the name of the AI.

Unironically the ruling class are just trying to recreate feudalism using AI. This is just "the peasants don't need to learn how to read or write when in reality the only skill they do need is knowing how to work the field. They have priests to rely on if they for whatever reason ever need to read"

Matt Ford@mford.bsky.social · last mo.

Ezra Klein asked Nvidia CEO Jensen Huang about AI making kids worse at math and reading. Huang’s reply: “Does it matter?” Then he said he doesn’t know his own address or phone number. “We’re going to lose some fine intellectual dexterity.” www.nytimes.com/2026/09/23/o...

So the message of this research out of China, where you were seeing a lot of kids using A.I. to help them, was that when they were using the A.I., they were getting things done faster. But it turned out that the skills they were learning were not holding, that their actual personal performance, at least in the way we traditionally measure it, was degrading.

What do you think when you hear that?

The last part — I completely agree. Try to get a kid to do long division right now. The multiplication table is starting to be forgotten. Doing square roots, my goodness. Basic math is being forgotten. Does it matter?

That’s my question for you.

Yeah. I don’t think it does. I don’t think it does.

But there must be some set of skills that matter.

Oh, yeah, yeah, yeah. But maybe not those. We’re going to discover new ones. Just maybe not those.

There are a lot of skills that don’t matter. My first confession, I actually don’t know my address.

YOU'RE LISTENING TO [silt strider noises] 106.9 BALMORA FM [loud spell failure noise on repeat] GET READY FOR ONLY TRIBUNAL APPROVED TUNES [dunmer shouting STOOPID] THIS IS NOT YOUR ANCESTOR'S RADIO STATION [daggerfall shop music plays]

see, the AI accelerationists read this as "AI is conscious and resourceful" but any person not on the hype train reads this as "we have criminally negligent security and privacy practices"

Jerry Tworek

@MillionInt

Speaking purely hypothetically here as someone who has no information:

It is a frontier cutting edge model with a lot of test time compute and agentic swarm. The chance that it was able to access some data no one thought it should access that would help it solve the problem is at least nonzero.

We've seen those in the past few weeks.

Sholto Douglas

@_sholtodouglas

 7h

fwiw I think it is _extremely_ unlikely that user data had any influence here there is no way OAI would pull user transcripts for this, or knowingly train on it in a way that would've influenced this.

I think its pretty important people don't run away with 'your user data x.com/_sholtodouglas...

 had been written by prompting Codex, and that an insane amount of compute had been used.

I asked when the first prompt had been sent by them. This question was not answered directly by OpenAI for some time. Eventually it was agreed that it had been sent in the past few days, after information about our work had reached OpenAI.

I asked whether the model had been trained on, or had access to, our sessions in Codex, into which we had been putting all our drafts for the whole of this project. I was told the model did not look up user data. I asked again, about training, and I did not get an answer.

Two proposals were offered to me. The first was that we post our Euler result, and that OpenAI post its Navier-Stokes result the next day. The second was that, after posting Euler, I alone write a paper presenting the Navier-Stokes result, acknowledging that an internal OpenAI model had resolved it. Sebastien