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 October 2026 links roundup collects commentary on AI-generated art and poetry translation, a bot’s win in the Metaculus Cup, and debates about marriage, suburbs and polyamory. The author warns that links may contain errors and says some claims and data in the roundup have not been independently verified.
Astral Codex Ten has published its October 2026 links roundup, a collection of short commentary on AI art, translation and forecasting alongside debates about marriage, suburbs and the history of polyamory. The post’s author cautions that the linked material has not been independently checked in every case, making the roundup a guide to discussions rather than a verified account of every claim it mentions.
Among the AI-related items, the author praises an AI-assisted image drawing on the style of Hieronymus Bosch, while acknowledging that its reliance on a human artist’s work complicates whether it should count as AI art. The post also describes an experiment translating the opening of the Chinese novel Romance of the Three Kingdoms into rhyming pentameter and links to an English rendering of the first 100 lines of The Odyssey in heroic hexameter. The author says these examples suggest promise for poetry translation, but makes clear that this is a personal assessment and that they do not speak Chinese.
The roundup reports that an AI bot named Laertes took first place in the Metaculus Cup forecasting competition in September. Citing an Economist account, the author says Laertes was developed by Jeffrey Liang, who estimated that he spent less than 150 hours and a few thousand dollars on computing and data. The post also names other high-performing bots and notes that AI startup Mantic had received $25 million in seed funding. The supplied material does not provide the funding announcement’s date or terms.
Other entries challenge or discuss claims in social commentary. The author disputes an argument that only 18% of marriages are non-miserable, saying the cited survey appears to show that 61% of respondents described their marriages as “very happy” and that the argument counted “pretty happy” marriages as miserable. The author says they checked that point but had not confirmed other alleged errors. The roundup also links to arguments about American suburbs and the origins of modern polyamory, and includes a historical item about Jeremy Bentham’s cookbook for a planned Panopticon prison.
AI Moves Into Forecasting and Translation
The selection presents AI as a tool being tested in activities that require more than producing text or images: forecasting outcomes and rendering poetry across languages while retaining a formal rhythm. The reported Metaculus Cup result offers a concrete competition outcome, while the translation examples are experiments described by the author rather than a formal evaluation. Together, they show why readers may want to distinguish performance in a specific task from broader claims about AI capability.
The roundup also illustrates how fast-moving technology stories sit alongside arguments that depend on careful reading of evidence. Its marriage entry points to a dispute over how survey categories were interpreted; its AI items raise questions about authorship, model behavior and what counts as meaningful performance. The author’s caveats make the post useful as a set of leads, but they also limit what can be treated as independently established from the roundup alone.
Popular right now
iRobot Roomba 105 Robot Vacuum with LiDAR Navigation
Hisense 65-inch U6 Pro Mini-LED 4K UHD Smart Fire TV
Apple AirPods Pro 3 Wireless Earbuds with Active Noise Cancellation
As an affiliate, we earn on qualifying purchases.
A Curated Mix With Caveats
Astral Codex Ten presents these entries as a monthly selection of links and personal reactions, not as a single reported investigation. The author explicitly says they have not independently verified every link. They estimate that commenters typically identify evidence that two or three links per roundup are wrong or misleading, and say they correct errors when they find them. That estimate describes the author’s experience; the source does not provide a systematic audit or specify a time period for measuring it.
This caveat is particularly relevant to the marriage discussion, where the author distinguishes between a source’s survey categories and another writer’s interpretation of them. It also applies to the post’s claims about AI systems and current events. The entries offer links and attributed descriptions, but the material supplied here does not include the underlying research papers, competition records or funding documents needed to verify every detail independently.
“I haven’t independently verified each link.”
— Astral Codex Ten author
Claims Still Need Independent Checks
The roundup does not establish that each linked report or claim is accurate. The author says some items may be corrected later and specifically notes that they had not confirmed all of the alleged errors in the marriage argument. The supplied text also does not include the underlying survey analysis, Metaculus results, AI system evaluations or Mantic’s funding announcement, so those details cannot all be independently assessed here.
Several interpretations remain open. The author’s enthusiasm about AI poetry translation is not based on a stated benchmark, and the post does not explain how the translations were evaluated. It also raises, without resolving, questions about AI-assisted art and whether language-model completions should be read as evidence of a system’s own views or desires. The source cautions against one interpretation of possible internal-email leakage but does not provide a technical investigation of the prompts or model behavior.
Corrections and Follow-Up Evidence
The author says they will correct errors as they notice them and highlight important corrections later. Readers seeking to verify individual items will need to consult the linked primary sources, including the survey, competition coverage and funding announcement. The post itself does not give a schedule for further updates or specify which links have already been checked.
For the AI claims, further evidence would include independent comparisons of translation quality, documented forecasting results across a defined period, and clearer information about how the systems were built and tested. Until such material is available, the roundup’s examples are best read as reported developments and personal commentary, not as proof of general AI capability.
Key Questions
What is the October 2026 post?
It is an Astral Codex Ten links roundup collecting brief commentary and links on AI, culture, social debates and other subjects.
What AI development does the roundup report?
It says the bot Laertes won the Metaculus Cup in September and identifies Jeffrey Liang as its developer, citing an Economist account. The supplied material does not include the underlying competition records.
Does the post prove AI can translate poetry better than people?
No. The author describes examples and argues that poetry translation may be a strong AI use case, but provides no formal benchmark or comparative study in the supplied text.
Are all the links and claims verified?
No. The author says they have not independently verified every link, and warns that some items may be wrong or misleading. They say they correct errors when identified.
What does the roundup say about the marriage statistics?
The author disputes an argument that only 18% of marriages are non-miserable, saying the cited survey appears to show 61% described their marriages as “very happy.” The author says they checked this point but had not confirmed other alleged errors.
Source: astralcodexten.com
