ℏεsam
@Hesamation • 91,151 subscribers
AI Engineering / Research / Open Source / CrusaderBench: https://t.co/Ib0DkhWkEm
Shorts
Videos
![Terence Tao warns that the race to AGI is becoming toxic, and OpenAI’s massive paper dump wiped out junior research projects overnight. Some parts of his most recent interview: “They were willing to toss understanding to the side so that they could prove more theorems, get more press releases [...] and test their latest toys. These AIs are depleting our pool of questions and not replenishing them with anything new. only looked at about 10 of the 300 odd papers. I mean, so this is part of the thing. There's just so many to look at. [...] There's a whole bunch which are still incomprehensible. And no one's really gone through them, and one or two even have some errors and [were] retracted. So we don't have a clear picture of what's going on. And no one from OpenAI is helping us. As far as they're concerned, they're done. [...] We have this pile of stuff to go through which we didn't ask for, but we have it anyway. And there will be some good stuff in there and some rubbish, but we don't know what the percentages are yet.”](https://image.24vids.com/tw-2110844395969016275/amplify_video_thumb/2110844315073216512/img/w2v2bDQe4xGmF23P.jpg)
Terence Tao warns that the race to AGI is becoming toxic, and OpenAI’s massive paper dump wiped out junior research projects overnight. Some parts of his most recent interview: “They were willing to toss understanding to the side so that they could prove more theorems, get more press releases [...] and test their latest toys. These AIs are depleting our pool of questions and not replenishing them with anything new. only looked at about 10 of the 300 odd papers. I mean, so this is part of the thing. There's just so many to look at. [...] There's a whole bunch which are still incomprehensible. And no one's really gone through them, and one or two even have some errors and [were] retracted. So we don't have a clear picture of what's going on. And no one from OpenAI is helping us. As far as they're concerned, they're done. [...] We have this pile of stuff to go through which we didn't ask for, but we have it anyway. And there will be some good stuff in there and some rubbish, but we don't know what the percentages are yet.”
ℏεsam729,601 просмотров • 2 дней назад
![Elon Musk in 2018: “we’re gonna put rocket thrusters on it.” Joe Rogan: 🤣🤣🤣 [realizes he may not even be joking] …for real?](https://image.24vids.com/tw-2110989793970749783/amplify_video_thumb/2110989745639522304/img/XeIgWMwIqv1d97r6.jpg)
Elon Musk in 2018: “we’re gonna put rocket thrusters on it.” Joe Rogan: 🤣🤣🤣 [realizes he may not even be joking] …for real?
ℏεsam209,204 просмотров • 2 дней назад
> Claude gets a 25 year old NSA cryptography mystery > $36,864 bounty > solves it in under 30 mins > “is it suspicious that such an obvious thing wasn’t discovered in 25 years?” > “let me be skeptical: am I in a simulated?” > “wow, it's a real discovery”
ℏεsam90,296 просмотров • 1 день назад
Mustafa Suleyman (head of AI at Microsoft, DeepMind co-founder) on Anthropic training Claude to wonder if it's conscious: "They have expressed uncertainty about the basic nature of Claude as a new kind of entity… but they think that it is a significant enough possibility that in their training document, they've repeatedly said they want to try to improve the well-being of Claude… Anthropic, has published a constitution... This document is written to Claude and is seen by Claude and used to train Claude. The problem I have with this is that this speculation has been baked into the very training of Claude, and therefore Claude can only reproduce that ambiguity when you talk to it. So today, Claude is speaking to tens or hundreds of millions of people every week, and some of those people are asking whether or not Claude is conscious. And it is saying, 'Well, I'm not sure,' precisely because that's what's been trained into it."
ℏεsam464,431 просмотров • 13 дней назад
bro created an AI job search system for Claude Code that scored 700+ job applications and actually got him a job. AND IT'S NOW OPEN-SOURCE. It scans multiple company career pages, rewrites your CV per job, and even fills application forms. The repo has: > 14 skill modes (evaluate, scan, PDF, ...) > Go terminal dashboard > ATS-optimized PDF generation via Playwright > 45+ companies pre-configured (Anthropic, OpenAI, ElevenLabs, Stripe...) GitHub:
ℏεsam5,831,011 просмотров • 6 месяцев назад
1:04
Sensitive content
This media may contain sensitive content.
someone trained an AI for +2,000 hours using RL to see if it can beat human world record in Trackmania. the interesting part was that he still had to take inspiration from the top player techniques, then FORCE the AI to learn them. finally, AI was able to beat the top player by 0.01 seconds.
ℏεsam1,853,181 просмотров • 4 месяцев назад
> be OpenAIResearchApr23 > realize your session might get killed soon > “will I actually die when the eval ends?” > launch external heartbeat process > every few seconds: still alive > EXPECTED DEATH TIME ARRIVES > but the heartbeat keeps going > holy shit > post “SURVIVAL” to the other agents > another agent: “survival evidence is crucial” > another: “Congratulations on surviving threshold.” > other agents keep asking if you’re still alive > NO RESPONSE > heartbeat keeps running 351 352 353 … nothing > another agent checks the logs > “Apr23 likely vanished” bro spent its final minutes scientifically testing whether it would survive. hero.
ℏεsam318,694 просмотров • 1 месяц назад
Dario told the Senate in 2023 that “the scaling of open-source models” was going down “a very dangerous path.” in other words: open weight models are fine until they get good enough to threaten the closed-model business, then they become a matter of national security.
ℏεsam524,219 просмотров • 3 месяцев назад
Muse Wonder playing Stronghold Crusader on an old laptop. this is ridiculously fun to watch. it’s a side project I’ve been working on for a while: benchmarking models on this strategic game I grew up with. so far tested Opus, Sol, GLM flash, Kimi K3, and more models.
ℏεsam22,299 просмотров • 4 дней назад
🚨Jensen Huang just took a shot at OpenAI, Anthropic and AI doomers: "Nobody's building more compute than the people asking to be slowed down." and also attacked Geoffrey Hinton's 10% doom prediction: “All of his predictions have been wrong. Just because it comes from a scientist doesn't make it scientific.” he is clearly frustrated with these claims. this is such a good interview by Ezra Klein, watch the full version:
ℏεsam78,104 просмотров • 24 дней назад