this is pretty much the only way to build product in AI models move so fast, that you have to be willing to throw everything out every few months to take advantage of what's possible
AI
-
Deep Learning Limitations: Researcher Demands Apology from Sama
By
–
Dear @sama,
— Gary Marcus (@GaryMarcus) 15 mars 2026
You owe me an apology. You have relentlessly, publicly and privately, attacked my integrity and wisdom since my 2022 paper “Deep Learning is a Hitting a Wall”.
But in your own way you have just come around to conceding *exactly* what I was arguing in that paper:… https://t.co/qww1bGjPbdDear @sama
, You owe me an apology. You have relentlessly, publicly and privately, attacked my integrity and wisdom since my 2022 paper “Deep Learning is a Hitting a Wall”. But in your own way you have just come around to conceding *exactly* what I was arguing in that paper: -
Scaling limitations in AI: experiments settle the debate
By
–
integrity and intellectual honesty are vital. and experiments, at the cost of tens of billions, have settled it. scaling isn’t working, and that’s why Sam is saying what he is saying. and it confirms that my intuitions were correct i am sorry you missed all that context.
-
Memory Gaps in AI: Compression, Association, and Forgetting
By
–
Well said. Compression, association, and forgetting—these three are indeed core capability gaps. In EverMemBench, we measured the cross-group accuracy dropping from 54.5% to 19.7%, which essentially means the "association" link is completely missing. I'm also optimistic about the
-
OpenAI’s Culture of Risk-Taking and Continuous Innovation at Codex
By
–
Working at OpenAI is fun because questioning everything and taking risks is part of the culture. Within Codex, the team asks itself how we could make it an order of magnitude better every few months and then sets most things aside to go and do it across the entire stack. Some examples were the Codex App and our first deployment of Cerebras inference with WebSockets. We are now well under way on the next bet and it’s making even our best engineers nervous as it’s at the edge of what’s possible today.
-
Deep Hitting Is Hitting: Research Paper Discussion
By
–
hilariously that was literally the point of my much-hated 2022 paper “Deep Hitting Is Hitting”
-
Incandescent vs. LED Lighting: Infrared Light and EU Regulations
By
–
Excited to Support Dr. Datta and his lab in their mission to push healthcare research in India! Codex India
-
Llama.cpp Alternatives Recommended for Multi-GPU Inference Setups
By
–
Keep in mind that once you get into concurrency and tensor parallelism (multi-GPU setups), you should be recommending inference engines other than llama.cpp
-
Network Protocol Security and Encryption Vulnerabilities
By
–
@jelmerdeboer deze man heeft goeie bio
-
Altman’s Architecture Claims Need Critical Examination
By
–
Misleading summary. Should be deleted.
— Gary Marcus (@GaryMarcus) 15 mars 2026
Altman doesn’t say a (known) new architecture is coming; he says he anticipates one will come someday.
PS: I also think we need something radical and new. In fact that’s what I’ve been saying for the last decade.
Excess focus on… https://t.co/qww1bGjPbdMisleading summary. Should be deleted. Altman doesn’t say a (known) new architecture is coming; he says he anticipates one will come someday. PS: I also think we need something radical and new. In fact that’s what I’ve been saying for the last decade. Excess focus on