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 Open Thread 456 is a weekly community post combining reader discussion with updates on meetups, election guides and AI-related projects. The post also invites people familiar with AI safety and grantmaking to complete surveys intended to help model how grants are evaluated.
The thread invites readers to post about any subject and points them to the community’s unofficial subreddit, Discord and bulletin board, as well as in-person gatherings. It reports meetup changes in Rockville, Northampton, Eindhoven, Great Barrington, Cairo, Portland and Sofia, but directs readers to the original post for details.
The author says the Boston ACX meetup group completed a Greater Boston general-election voting guide and that additional guides may follow. Work on a Bay Area guide was still under consideration. The post also said San Francisco mail-in ballot registration was due by the end of that Monday and identified Scott Wiener for Congress as the group’s main San Francisco endorsement. These are the post’s statements; the supplied material does not independently verify election information.
For the grantmaking project, the author asks people versed in AI and grantmaking to complete a survey rating roughly 80 AI safety accomplishments. A second, alternate survey asks respondents to predict how the author would answer the first one. The post says this comparison is meant to distinguish differences in worldview from random variation in how people assign weights.
How the Survey Could Inform Grantmaking
The survey is an attempt to gather structured judgments about which AI safety achievements deserve greater weight when evaluating grantmaking. Its results could help inform a model of the decision-making process, though the post describes work in progress rather than a finished or automated system.
The alternate survey addresses a separate methodological issue: whether differences between responses reflect distinct views about the value of accomplishments or noise in assigning scores. That distinction matters if the author uses responses to build a model of grantmaking preferences. The post does not say how responses will be analyzed, how many participants are sought, or whether results will influence actual funding decisions.
The other announcements also have practical relevance for readers: meetup changes affect local participation, while voting guides and registration reminders concern election activity. The post’s remaining project notes point to experiments in prediction markets and user-rated notes, but give readers limited information about their operation or reach.
AI safety accomplishment rating survey
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
A Weekly Forum With Project Updates
Open Thread 456 is presented as a recurring community discussion post, not as a standalone research paper or formal announcement from an institution. Alongside the open invitation to discuss anything, the author uses the thread to circulate local event information and requests for participation in ongoing work.
The post describes the AI grantmaking effort as continuing and links it to previous efforts, but the supplied source material does not detail those earlier projects. It also previews a Bay Area ACX meetup for Saturday, October 24, at 1 p.m. at Mox, 1680 Mission St. in San Francisco, saying a separate post may follow.
Two additional projects are mentioned. The post says Manifold Markets co-founder and grantee Stephen Grugett announced MNX, described as a real-money prediction market focused on AI. It says users can trade futures representing AI company stocks, bet directly on company revenues and make predictions tied to Epoch Capabilities Index targets. The author also highlights Common Notes, a Goodheart Labs project described as a system for community notes across the web, using AI-generated notes, user contributions and voting, with a browser extension.
““I am continuing to work on modeling (and maybe one day automating) some parts of AI grantmaking.””
— Astral Codex Ten author, in the Open Thread 456 post
As an affiliate, we earn on qualifying purchases.
Survey Results And Project Details Pending
The post does not report survey results, identify a target number of respondents, specify a closing date or explain how the answers will be used in practice. It is also unclear whether the AI grantmaking model will be published, used to guide funding choices or eventually automated.
Several time-sensitive details need checking against current information. The post refers to a mail-in registration deadline as “today (Monday)” and lists an October 24 meetup, but the supplied material does not give a publication year. It also says meetup arrangements changed in seven cities without providing those changes in the excerpt.
The descriptions of MNX and Common Notes come from the open thread. The source does not provide independent verification of the services’ features, availability, market rules or user base. The comparison of Anthropic-related pricing on Polymarket and MNX is likewise presented as the author’s observation, without a specified measurement time or supporting data in the supplied text.
vote-by-mail ballot registration kit
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Upcoming Surveys And Local Meetups
Readers interested in the grantmaking research can take the main survey and, if they have time, the alternate prediction survey linked in the original post. The author says additional election guides may be published in the coming week and that a separate announcement about the Bay Area meetup may follow.
The October 24 San Francisco gathering is listed for 1 p.m. at Mox, while readers in the other named cities are directed to the post for meetup changes. The thread also points readers to MNX and Common Notes for more information about those projects. The source gives no schedule for the next survey update or further details on the grantmaking model.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is Open Thread 456?
It is a weekly Astral Codex Ten post inviting open discussion and sharing community announcements, including meetup, election-guide and project updates.
What does the AI grantmaking survey ask people to do?
The main survey asks people familiar with AI and grantmaking to assign utility weights to about 80 AI safety accomplishments. An alternate version asks respondents to predict how the author would answer the main survey.
Why is there an alternate survey?
The author says it is intended to help distinguish differences in respondents’ worldviews from random variation in how they assign weights. The post does not report results or explain the analysis in further detail.
When and where is the Bay Area meetup?
The post lists a meetup for Saturday, October 24, at 1 p.m. at Mox, 1680 Mission St. in San Francisco. The year is not specified in the supplied material, so readers should check the original post for current details.
What are MNX and Common Notes?
The post describes MNX as a real-money prediction market focused on AI-related companies and measures. It describes Common Notes, from Goodheart Labs, as a system for notes across websites using AI, user-written contributions and voting, with a browser extension.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
