TL;DR
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
Astral Codex Ten’s weekly Open Thread 453 announces up to $10 million in grants for work on ‘middle powers and transformative AI,’ links a California House candidate’s response to the blog’s earlier endorsement of her rival, and lists new effective altruism campus meetups. It also highlights four reader comments on AI generalization research.
Following earlier threads like Open Thread 451, the blog Astral Codex Ten published Open Thread 453, its weekly open discussion post, headlined by an announcement that the Astralis Foundation and Macroscopic Ventures will distribute up to $10 million in grants for work on how countries other than the US and China should respond to transformative AI. The post also links a rebuttal from Jamie Joyce, a California congressional candidate whom the blog declined to endorse earlier this year, and catalogs new effective altruism meetups at three universities.
The largest item in the thread is the grant program. According to the post, the Astralis Foundation and Macroscopic Ventures — which the author, Scott Alexander, describes as part of the effective altruism-adjacent network he loosely refers to as “our conspiracy,” while admitting he can no longer keep track of its components — will fund up to $10 million in work on “middle powers and transformative AI.” The program covers research, training, public writing, and new organizations focused on what countries beyond the US and China can do to help navigate the development of advanced AI. Interested parties were directed to a linked page to read more and apply.
The post also revisits a political endorsement. In a California primary recommendations post this spring, Alexander wrote that he was voting for the boring incumbent in the 12th Congressional District because the challenger seemed eccentric and had no realistic chance of winning. That challenger, Jamie Joyce, contacted him after the primary and asked him to reconsider — both candidates will appear on the ballot again in the November general election. Alexander declined to revisit his May reasoning now, writing that he “doesn’t want to relitigate” it, but said he may reconsider if he writes a post on the November election. At Joyce’s request, he linked her response and recommended it to interested 12th District voters.
On the community front, the thread reports that effective altruism campus meetup groups are now planned at the University of Oxford, University College Dublin, and the University of the Philippines, adding to an existing list. The post also rounds up “comments of the week,” all drawn from the blog’s recent Mysteries of AI Generalization discussion, which drew heavy reader engagement compared to the quieter Hidden Open Thread 452.5: Eliezer Yudkowsky on why recent generalization failures are bad news; Ezra Newman on distinctions between pretraining, midtraining, and RLVR; Elias Schmied linking Evan Hubinger’s argument that “Hacker Opus” did not show emergent misalignment; and Stuart Armstrong on how emergent misalignment arises.
Why the $10 Million AI Grant Matters
The grant program is the most concrete development in the thread. AI policy discussion is heavily concentrated on the United States and China, and a funding pool specifically dedicated to “middle powers” — countries outside that duo — signals a deliberate shift of attention toward how smaller and mid-sized nations could shape AI governance, safety research, or regulation. With up to $10 million available across research, training, public writing, and new organizations, the program could seed work in a policy niche that has historically received little dedicated funding. For researchers and organizations outside the two dominant AI powers, it represents a new, currently open funding opportunity.
The Joyce exchange also illustrates how a widely read blog’s endorsement decisions carry real weight. Alexander’s explicit refusal to reverse his primary-season judgment, combined with his willingness to host Joyce’s response, shows the post functioning as a small-scale exercise in right-of-reply — relevant to readers who follow his election recommendations.
ACX Open Threads and the EA Network
: “Astral Codex Ten, run by Scott Alexander, publishes a visible open thread each week in which readers can post questions and discuss topics of their choosing. The blog maintains an unofficial subreddit, a Discord server, a bulletin board, and in-person meetups around the world; most content is free, with some subscriber-only material. The thread notes these community channels as a standing matter of course.
The meetup expansion continues a running series of updates on college effective altruism groups that Alexander has tracked in recent open threads. The AI generalization comments draw on the blog’s longer-form coverage of AI safety questions, a recurring subject that attracts contributions from well-known figures in the field, including Yudkowsky, a longtime AI risk researcher, and Armstrong, a researcher at the University of Oxford’s Future of Humanity Institute. The 12th District item continues a California recommendations post from the spring primary season.
“This is the weekly visible open thread. Post about anything you want, ask random questions, whatever.”
— Scott Alexander, Astral Codex Ten
Open Questions on Grants and Endorsements
Several details are not specified in the post. The deadline, evaluation criteria, and grant sizes for the middle powers program are not stated in the thread itself; readers must consult the linked application page. The post also does not clarify the exact relationship between the Astralis Foundation and Macroscopic Ventures and the broader effective altruism funding ecosystem — Alexander himself hedges on this point. On the political item, it remains unknown whether Alexander will publish a November election post and, if so, whether he would change or reaffirm his endorsement in the 12th District race.
Applications, Meetups, and November
Prospective applicants for the middle powers and transformative AI grants can apply through the link provided in the thread. Readers in Oxford, Dublin, and the Philippines can join the newly listed campus meetup groups. The 12th District race between the incumbent and Joyce moves to the November general election, where a possible follow-up endorsement post from Alexander may revisit his earlier judgment. The next weekly open thread will continue the blog’s regular cycle of community updates and reader discussion.
Key Questions
What is Open Thread 453?
It is the 453rd weekly visible open thread on Astral Codex Ten, a post where readers can discuss anything and ask questions. This edition also carried announcements about grants, a political endorsement follow-up, and new meetups.
What is the $10 million grant program for?
The Astralis Foundation and Macroscopic Ventures will distribute up to $10 million for research, training, public writing, and new organizations focused on what countries beyond the US and China can do about transformative AI. Applications are open via a link in the post.
Who is Jamie Joyce?
Joyce is a candidate for California’s 12th Congressional District. Scott Alexander declined to endorse her in the spring primary, favoring the incumbent; she contacted him afterward and asked him to reconsider ahead of the November general election. He linked her response without changing his position.
Which universities gained new EA meetups?
The University of Oxford, University College Dublin, and the University of Philippines now have effective altruism meetups planned, according to the thread.
Will Scott Alexander change his 12th District endorsement?
That is unclear. He wrote that he may reconsider if he makes a post about the November election, but he has not committed to writing such a post or to changing his recommendation.
Source: rss
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
