talking to an OpenAI friend be like: > “How was dinner?” > “It’s a good model, sir.”

All of these LLMs talk the same because: - there's only one Internet - Transformer is all you need - everyone is doing pre-training → supervised fine-tuning → RL - all the “secret sauce” leaks out between the big labs at SF restaurants

DeepMind and OpenAI ranked 27th at IMO. I feel this will be the last year human students beat AI at IMO. Big congrats to the cracked kids! All 4 U.S. medalists who beat AI seem to be Chinese-American. Didn’t expect Alexander Wang to take Gold, especially when he is leading Meta’s Superintelligence Labs. (ofc a joke!)

This wins my respect.

Over 10 OpenAI researchers turned down $300M offers from Zuckerberg. Guess they're the "missionaries" Sam talked about. “Meta still doesn’t have a chief scientist for its lab” Wait, wasn't that supposed to be Yann LeCun?

What is not a sign of intelligence but people think it is?

Playing it safe guarantees mediocrity. The biggest risk is not taking any risk.

Heard GPT-5 is imminent, from a little bird. - It’s not one model, but multiple models. It has a router that switches between reasoning, non-reasoning, and tool-using models. - That’s why Sam said they’d “fix model naming”: prompts will just auto-route to the right model. - GPT-6 is in training. I just hope they’re not delaying it for more safety tests. :)

Heard GPT-5 is imminent, from a little bird. - It’s a mixture of multiple models with a router that switches between reasoning, non-reasoning, and tool-using models. - That’s why Sam said they’d “fix model naming”: prompts will just auto-route to the right model. - GPT-6 is in training. I just hope they’re not delaying it for more safety tests. :)

Heard Zuck poached 4 more OpenAI researchers, including some behind the open-source model. how deep are Zuck’s pockets?