toucan (@distributionat) / X

toucan

4,620 posts

toucan

@distributionat

toucan beaks are models of lightweight strength • prev @AnthropicAI @scale_AI

Joined December 2021

Pinned
toucan
@distributionat
Jun 14
FAILURE OF IMAGINATION—Sure, everybody is clearly racing for recursive superintelligence right now. And the path forward seems so obvious: just automate software engineering, ML engineering and research, and hook it back up to itself. But if automating coding was so obvious, why
2.3K
toucan
@distributionat
Jun 30, 2025
Nerds have been complaining that athletes are overpaid for decades. Finally, the nine figure nerd trading market has emerged. Take that, LeBron Jameson.
RSC ☀️🌲
@silver__tsuki
Jun 30, 2025
On the left is Ronaldo, Real Madrid spent 80M $ to sign him from Man United On the right is Jiahui Yu, Meta paid 100M $ to sign him from OpenAI
293K
toucan
@distributionat
Nov 2, 2025
From Ilya’s deposition— • Ilya plotted over a year with Mira to remove Sam • Dario wanted Greg fired and himself in charge of all research • Mira told Ilya that Sam pitted her against Daniela • Ilya wrote a 52 page memo to get Sam fired and a separate doc on Greg
17M
toucan
@distributionat
Feb 3, 2025
If you get $500 of value per query, then in the first month of a Pro subscription you net $49,800 with your 100 queries. Then in month 2 you can buy 249 new subscriptions, and you net $12,450,000. By month 4 you make $778,125,000,000 and you can buy OpenAI.
Sam Altman
@sama
Feb 3, 2025
50 cents of compute for 500 dollars of value
447K
toucan
@distributionat
May 19, 2025
Replying to @khoomeik @MatternJustus and @aryaman2020
never never never kill kill kill yourself yourself yourself
153K
toucan
@distributionat
Jun 10, 2025
zuck is offering individual researchers the equivalent of Cursor's ENTIRE series B, that's how badly they're bleeding talent
448K
toucan
@distributionat
May 28, 2025
“the head of RL” at anthropic has gone to openai and the “head of RL” at openai has gone to anthropic you can’t make this stuff up
166K
toucan
@distributionat
Sep 15, 2025
Related: men underrate how much cognitive labor goes into skincare and how sophisticated women are about it. Getting into skincare (identifying unwanted skin conditions and fixing them) is one of the best ways to train empirical science skills. It's a practical science.
Patrick Collison
@patrickc
Sep 14, 2025
Observing some people close to me with chronic health conditions, it's striking how useful Reddit frequently ends up being. I think a core reason is because trials aren’t run for a lot of things, and Reddit provides a kind of emergent intelligence that sits between that which any
281K
toucan
@distributionat
Nov 2, 2025
Replying to @distributionat
• OpenAI is paying Ilya’s legal bills
390K
toucan
@distributionat
Nov 4, 2025
living in nyc puts the question of whether you’re ultra hot to the test or if you were just locally optimal hot in the pond you were previously in typically moving out after a few years is a silent admission you’re in the latter
110K
toucan
@distributionat
Jan 29, 2025
r1 is smarter in chinese than english. it has a different personality between the two languages. in english it’s kind of meh. in chinese it’s way smarter and cogent. i know there’s a lot of hype around it but i think twitter is underrating it.
83K
toucan
@distributionat
Jul 7, 2025
If I had $1B to put into AI, I would take the obvious bets on big labs, cloud, and energy. I would spend the high risk high reward bucket on robotics; infra and software built for when most internet traffic is AI (happening soon); and consumer AI.
147K
toucan
@distributionat
Jun 10, 2025
The plan? We're going to scale ai by BUYING Scale AI.
45K
toucan
@distributionat
Apr 21, 2025
this is AGI
39K