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 published Open Thread 453, its weekly visible open thread. The post points to up to $10 million in grants for work on middle powers and transformative AI, new effective altruism campus meetups in three countries, and a follow-up on a California House race the blog covered in the spring.
The blog Astral Codex Ten published Open Thread 453, its weekly visible open thread where readers can post questions and discussion on any topic. Alongside the open discussion, the post carries several announcements, including up to $10 million in grants for work on how countries beyond the US and China should respond to transformative AI, new effective altruism campus meetups at three universities, and a follow-up to the blog’s spring coverage of a California congressional race.
The largest concrete item in the post is a grant program announced by the Astralis Foundation and Macroscopic Ventures, which the author described as organizations he believes are part of the rationalist and effective altruism community network, while noting he can no longer keep track of its full structure. The two organizations say they will distribute up to $10 million for work on “middle powers and transformative AI” — defined in the announcement as “research, training, public writing and new organizations focused on what countries beyond the US and China can do to help navigate” the AI transition. The post links to a page where interested applicants can read more and apply.
The thread also updates the blog’s running list of college effective altruism meetups, adding newly planned groups at the University of Oxford, University College Dublin, and the University of the Philippines. Readers are directed to a maintained list for details on times and locations.
On the political side, the post revisits the author’s spring California primary ballot recommendations, in which he said he was voting for the “boring incumbent” in the state’s 12th Congressional District because the challenger appeared eccentric and unlikely to win. The challenger, Jamie Joyce, contacted the author and asked him to reconsider, and both candidates will face each other again in the November general election. The author declined to relitigate his May assessment but said he may revisit the race if he writes a post on the general election, and he linked to Joyce’s response for interested 12th District voters.
Why Middle-Powers AI Funding Matters
The $10 million grant announcement reflects a growing focus within effective altruism-aligned philanthropy on the international dimensions of AI policy. Rather than concentrating only on US or Chinese AI development, the program explicitly targets what smaller and mid-sized countries can do during a period of rapid AI progress — a framing that could fund research, journalism, and new organizations outside the two dominant AI powers.
For readers of the blog, the open thread also functions as a coordination point: meetup announcements spread local group formation across multiple countries, and the election follow-up shows the blog’s ballot-recommendation posts generating direct pushback from candidates. Both dynamics illustrate the community’s role as a network for organizing as well as discussion.
How ACX Open Threads Work
Astral Codex Ten, run by Scott Alexander, publishes a visible open thread each week in which readers can raise any topic or question. The blog maintains an unofficial subreddit, a Discord server, a bulletin board, and a network of in-person meetups around the world. Most of the site’s content is free, with some posts restricted to paying subscribers.
The spring primary post referenced in the thread was part of the blog’s periodic election recommendation series. The referenced “Mysteries of AI Generalization” post, from which the comments of the week were selected, examined how and why AI systems generalize — or fail to — beyond their training data, a recurring topic of debate in AI safety circles.
“This is the weekly visible open thread. Post about anything you want, ask random questions, whatever.”
— Astral Codex Ten, Open Thread 453
Grant Details and Election Plans Undecided
Several details are not specified in the post. The exact deadline, eligibility criteria, and grant sizes for the Astralis Foundation and Macroscopic Ventures program are not stated in the thread itself; interested parties are directed to the linked application page. The relationship between the two organizations and the broader network is described only loosely by the author.
Whether the author will write a follow-up post on the November general election in California’s 12th District — and whether he would change his recommendation — remains open. The dates and formats of the newly announced campus meetups at Oxford, University College Dublin, and the University of the Philippines are likewise left to the linked list.
Applications, Meetups, and a November Post
Applicants can submit proposals to the middle-powers AI grant program through the application page linked in the post, with up to $10 million available. The college EA meetup list should continue to grow as the academic term proceeds, and readers can check the pinned list for events at Oxford, University College Dublin, and the University of the Philippines.
The next installment in the numbered open-thread series will follow in roughly a week, and a possible ballot-recommendations post for the November general election — including a possible revisiting of the 12th District race between the incumbent and Jamie Joyce — may follow in the fall.
Key Questions
What is Open Thread 453?
It is the weekly visible open thread on the blog Astral Codex Ten, where readers can post about any topic or ask questions. This edition also carried announcements about grants, meetups, and a California election follow-up.
Who can apply for the $10 million AI grants?
The post does not specify eligibility details. The program from Astralis Foundation and Macroscopic Ventures funds “research, training, public writing and new organizations” focused on middle powers and transformative AI; readers are directed to the linked application page for details.
What are the new effective altruism campus meetups?
According to the post, new meetups are planned at the University of Oxford, University College Dublin, and the University of the Philippines, with details available in the blog’s maintained list.
Did the blogger change his endorsement in the California 12th District race?
No. The author said he does not want to relitigate his May primary recommendation but may reconsider if he writes a post about the November general election. He linked to challenger Jamie Joyce’s response for interested voters.
What were the comments of the week about?
All were drawn from the blog’s “Mysteries of AI Generalization” post, covering topics such as why recent generalization failures matter, distinctions between training stages, and how emergent misalignment arises — with contributions from Eliezer Yudkowsky, Ezra Newman, Elias Schmied, and Stuart Armstrong.
Source: rss
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
