Dust collection systems - a surprising amount of my cognitive cycles these days thingsilearned.com/things/dust-co…

Introducing Index – the new planning tool for Product Managers on Linear. ➡️ index.inc Product leaders have long deserved a modern experience for product planning and discovery. Something powerful, flexible, and integrated with the latest tools. That’s why we

Aider with Claude Sonnet makes it almost easier to create the app i want than find one - even if I know it exists thingsilearned.com/things/buildin…

Does every new js/ts/node project just require 4 days of configuration hell and rewrites or am I doing something wrong?

The LessWrong + Lighthaven fundraiser is about to end, with 24 hours on the clock. We've raised almost $2,000,000, all from people donating less than $200k, and an average donation size of $3,000.

We all have an obligation to society to communicate the reality of the situation to them. Stop it with vague schizoposting, stop downplaying what you really believe. Warn people that if this goes wrong, they're dead. Their children are dead. Humanity is done for.

Very relevant piece by Kevin Roose in The New York Times, 3 points that particularly resonate with me: 1⃣ AGI's arrival raises major economic, political and technological questions to which we currently have no answers. 2⃣ If we're in denial (or simply not paying attention), we could

I've had this document started and mostly empty mulling in the back of my mind for over a decade. With the pace of things I figured I needed to get it out. Today I published my letter to a future Superintelligence thingsilearned.com/things/letter-…. .

Containment requires foresight into possible escapes. But when the prisoner is smarter than the jailer, foresight collapses — the gap in intelligence is by definition unknown unknowns.

A riddle: What is safe and incredibly powerful yet no one can control it?

I hate so many things about slack, but their required install process to simply "add a new helper tool" is unacceptable. It requires you to put in your password multiple times (perhaps for multiple tools?), and will not let you cancel. Enraging

I published a helper tool for cleanly making complex prompts github.com/davefowler/llm…

It is difficult to make the world understand the danger posed by ASI. Much of that difficulty is because of human psychology, but some of the difficulty is manufactured by the AI industry, which uses Tobacco industry playbook tactics to prevent interference