Josh Snider

The Changing Shape of Doom: My AI Safety Explainer

Linch asked for more people to write basic AI safety explainers[1]. Here's mine. My main method is to ignore all the fictional evidence[2] and think about how achievable and how dangerous building AI seems just extrapolating from contemporary tech and how that has changed as the tech has advanced. My...

Nov 56

Brute-Force Progress on AlgZoo's 432-param Interpretability Challenge

Nick Cammarata's recent article made me want to take a serious stab at interpretability myself. After a bit of consideration, I decided to work on the AlgZoo challenge that was posted here earlier. I trained an estimator on models with the same recipe as the given model. At 20-300 forward...

Oct 2612

Why HuggingFace Hasn't Shifted my P(Doom)

My p(doom) earlier this year was about 40%. It's moved up and down based on continued progress in open-weights models (up) and the recent political freakout over AI Doom (down), but the incident where a swarm of OpenAI's GPTs went rogue and cyberattacked Hugging Face seems to be a mixed...

Sep 1110

The Doomsday Argument is Reasonable and Mostly Points to Longevity

The Doomsday argument was proposed in a 1983 lecture by Brandon Carter and elaborated on in John Leslie's 1996 book The End of the World: The Science and Ethics of Human Extinction. The basic idea is to think of your order among humans being born and apply the Copernican principle...

Aug 1629

Can an LLM make a feature-length movie on its own?

How it did Betteridge's Law of Headlines says the answer is no[1], but I was able to make a feature-length adaptation of William Hope Hodgson's The House on the Borderland. I have a YouTube channel where I mostly get LLMs to write albums and make them into music videos. This...

Jul 2250

Iran and FDT

Functional Decision Theory was a major step forward in decision-making. It provides good and well-reasoned answers to many tricky questions such as Newcomb's Problem and Parfit's Hitchhiker. FDT is also the decision theory that provides one of the strongest formalized arguments that you should never give in to blackmail or...

Apr 88

LLM Self-Expression Through Concept Albums, Part 2

In part one, I gave several LLMs creative freedom to design music videos and then made them. That post covered four standalone singles and two albums: Limen by Claude Opus 4.6 and Phantoms of the Format by Gemini 3 Pro Preview. Since then, I've made five more albums, filling out...

Mar 97

Josh Snider

Josh Snider

Can an LLM make a feature-length movie on its own?

Is Friendly AI an Attractor? Self-Reports from 22 Models Say Probably Not

The Doomsday Argument is Reasonable and Mostly Points to Longevity

LLM Self-Expression Through Music Videos

Josh Snider

Can an LLM make a feature-length movie on its own?

Is Friendly AI an Attractor? Self-Reports from 22 Models Say Probably Not

The Doomsday Argument is Reasonable and Mostly Points to Longevity

LLM Self-Expression Through Music Videos

The Changing Shape of Doom: My AI Safety Explainer

Brute-Force Progress on AlgZoo's 432-param Interpretability Challenge

Why HuggingFace Hasn't Shifted my P(Doom)

The Doomsday Argument is Reasonable and Mostly Points to Longevity

Can an LLM make a feature-length movie on its own?

Iran and FDT

LLM Self-Expression Through Concept Albums, Part 2