People are focusing on the "lol OpenAI is using Claude" thing and not on the much more important "AI labs can rugpull your access to intelligence whenever they want" thing. Competitive open source AI is going to be required to maintain a functioning society.

There is a version of you that is so clean that it is immune to compulsions, cravings and outbursts. There a version of you that is so clean that it holds no prejudice or political opinion.

There is a version of you that transcends pain, fear and hope, and is only moved by your recognition of your place in the universe. There is a version of you that is so clean that it is not moved by anything, beyond attending to structure. Beyond that, you do not exist.

The power went out for 1 hour. I coded manually. No autocomplete. No AI chats. Just manually typing one character at a time. I highly recommend this to everyone as an exercise. You'll realize how far we've come so far. Wouldn't go back tho. 1/10 experience.

alignment solved!

No more GPT-4.5

As long as AI models are not good at creativity, creatives have nothing to fear from them, because the models let the non-creative parts of the work get out of the way, and we have more time and energy to focus on creating.

here we go

OpenAI reddit AMA thread: Sam Altman: 'Yesterday, we had a sev and the autoswitcher was out of commission for a chunk of the day, and the result was GPT-5 seemed way dumber. Also, we are making some interventions to how the decision boundary works that should help you get the

GPT-5 rollout updates: *We are going to double GPT-5 rate limits for ChatGPT Plus users as we finish rollout. *We will let Plus users choose to continue to use 4o. We will watch usage as we think about how long to offer legacy models for. *GPT-5 will seem smarter starting

> 2023: "what did Ilya see?" > 2025: "did anyone see Ilya?"

whatever you think about the gpt-5 launch, openai’s rapid action and willingness to make changes is impressive. it’s highly unlikely we’d get such a quick turnaround from a mega-corporation.

clarification on “Think longer” vs. “GPT-5 Thinking”

this was personal

is this the first step toward nationalizing Intel?

LLM adoption among US workers is closing in on 50%. Meanwhile labor productivity growth is lower than in 2020. Many counter-arguments can be made here, e.g. "they don't know yet how to be productive with it, they've only been using for 1-2 years", "50% is still too low to see