Open Thread 311
...
This is the weekly visible open thread. Post about anything you want, ask random questions, whatever. ACX has an unofficial subreddit, Discord, and bulletin board, and in-person meetups around the world. 95% of content is free, but for the remaining 5% you can subscribe here. Also:
1: Last week I posted a subscribers-only book review of Cyropaedia. Several people commented that two days earlier, the Substack Mr. and Mrs. Psmith’s Bookshelf also reviewed the same book (the commenters politely omitted “and did a better job”). Highly recommended, whether you read my version or not.
2: Last year I bet that AI art generators would be able to handle some tough compositionality requests before 2025. There was widespread speculation that the latest generation (eg DALL-E2) had won this bet. Edwin Chen has tested them out, and says they’re not quite there.
3: The PIBBSS Fellowship asks me to remind you of their existence. They’re a 3-month fully-funded program in AI alignment. They take PhDs and postdocs from other fields "such as evolutionary bio, neuroscience, dynamical systems theory, economic/political/legal theory, and more" and help them work on a project "at the intersection of your own field and AI safety, under the mentorship of experienced AI alignment researchers". Learn more / apply here before the February 4 deadline.