I must be among an extremely small group of people (n=1?) that have both 1) trained a frontier LLM and 2) designed and synthesized custom viruses in a lab with my own two hands. And I think that the takes on AI killing us all by creating dangerous viruses is total bogus.↗

There are two ways AI progress could go very badly and that we must avoid. First, we could lose control of the future to AI. This is unacceptable; we are unapologetically on Team Humanity, and AI must always serve people. To ensure that, we need ways to ensure that alignment and…↗

i cannot imagine a bigger bag missed than being wrong about where the money is gonna land from superintelligence can you imagine explaining to your LPs how you missed that the spin you would have to pull↗

Here's the link to it :з 30desk.art↗

Wild 24 hours for AI and lots of different proposals have been made. TLDR; the only *tangible* new fact is that OpenAI and Anthropic are going to have embedded 3rd party evaluators from unknown organizations with Dario floating METR as a possibility. Having 3rd party evaluators…↗

Making Startups Powerful: paulgraham.com/powerful.html↗

Morning Bathrobe Rant: Rethinking Harnesses.↗

On macOS and wanna see something interesting? Ask your agent: "look at ~/Library/Application Support/Knowledge/knowledgeC.db and tell me some interesting facts" I had no idea this stuff was all getting logged and it can infer a lot about your activity.↗

for the skeptics in government and elsewhere: “pacing the frontier” will compress the margins of the frontier labs. it is a heavy cost imposed asymmetrically on model developers with the strongest AIs in America. by its nature, it would be a terrible regulatory capture tactic↗

An alternative hypothesis: - model performance is plateauing - compute is getting much more expensive - AI data centers are massively unpopular - slowing down AI development is a way to explain slowing progress, reduce spending, and try to regain some goodwill. - This is an effo…↗

When you want Astra to review snapshots of your app, ask it to collage them into a contact sheet to minimize the number of image read calls it needs to do. You’ll save a ton of context usage This works for fable too!↗

Anyone can get my friend’s real SSN from Claude I'm Terrified of my data in training sets. Emails in databases. iMessages harvested. Cant trust the cloud as AI is crazy good at hacking So I built Underdog for myself: on-device AI OS. Capable & 100% local. now my friends love it↗

Sam Altman told me yesterday there will be no IPO for OpenAI in 2026, that it would be “ill timed” given all they have to work on with alignment, control and safety fortune.com/2026/09/12/sam…↗

GPT-6 Astra by our community We shipped GPT-6 Astra. Builders took it from there: --- Explore 2,234 modeled anatomical parts in 3D: [Embedded Tweet: x.com/i/status/2096221988763173186] --- A Manhattan recreation built in Unreal Engine: [Embedded Tweet: …↗

Visual Diffs! For design AND code. Real-time look at me building @OpacityHQ! We have our OWN version control that has nothing to do with Git. Y'all, it's time to move off Github and Figma (genuinely love them both, but there's a new kid on the block) and onto Zoah!↗

@rcx86 If only I had thought to train it on StackOverflow instead I think it would have clicked for me that LLMs can be promptable, general purpose Q&A engines. Ah well.↗

i'm not procastinating, you are: wishlist-lego.vercel.app↗

I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we've had at OpenAI in recent weeks. Committing to having independent evaluators with employee-like access is a great idea, and we will do the same. We'll have more to share soon.↗

We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so. Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our s…↗

Rune Icons is live. 900+ icons, each in 5 styles: outline, duotone, fill, pixel, and glass. You can reshape any path in the browser and copy it as SVG or JSX. Free, open source, no sign-up. runeicons{.}com↗

i scraped 85 million birthdays and used astra to make a website that tells you how rare your birthday is. birthdaytraffic.com↗

Your pet has a job now. Pets help you keep track of your chats while you’re away from the desktop app. Now, you can start a new chat right from your pet, too. Not a pet person? Meet Mini: the same shortcuts and updates, taking up less space on your screen. Pick your pet—or go…↗

Givemeanode is now SF Autoresearch, a new product from SFC. We built the scalable, resilient platform for agent-driven machine learning research. The current moment is important and we want to help you take it seriously. sfcompute.com/news/sfc-autor…↗

It's here! Control any terminal session from your team from any device. Shell.online V2 gives you one unified view of every agent your org is running, and a live keyboard into each of them. So excited for this 1/👇👇👇↗

My YC batch wrapped up yesterday & I wanted to share a few of the companies I was close with and think you should check out! @hoplite_sh - cloud agents with a beautiful design - their onboarding flow is incredible and gets you up and running super quickly @try_glen - organizati…↗

BREAKING: 40 British MPs have signed a letter to Prime Minister Andy Burnham calling for a ban on the development of superintelligent AI.↗

If you want to play around with an early thing: you can now give us money and we give you tokens in return. radius.earendil.com↗

happy friday! i did some interaction work rauno.me/notes/5↗

better benchmark scores don't tell you much about how well a model does on your real work that's why for the last 3 years @every we've done vibe checks on new models: long-form reviews based on hands-on testing on each model for our real work now we're doubling down, and getti…↗

New in Claude Code: claude plugin eval See what value your plugin is adding, or if it needs more work. You can create test cases, run your plugin or skill against those test cases, score those runs, then run each case again without the plugin to see the differences.↗