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 its October 2026 links roundup, covering AI-generated art and translation, marriage statistics, and a bot’s Metaculus Cup win. The author says individual links have not been independently verified and that commenters often identify errors in these posts.
Astral Codex Ten has published its October 2026 links roundup, a collection of commentary on topics including AI-generated art and translation, marriage statistics and a bot’s win in a forecasting competition. The post is a curated commentary piece rather than a single reported investigation, and its author cautions that the links have not all been independently verified.
The roundup ranges widely. It opens with a note about Jeremy Bentham, the 18th-century philosopher, and a cookbook he reportedly wrote for use at the Panopticon prison. Other entries discuss an AI-generated image inspired by Hieronymus Bosch and an experiment rendering passages from the Chinese novel Romance of the Three Kingdoms in rhyming English pentameter. The author says the translation looks impressive at a glance but adds that they do not speak Chinese, so cannot evaluate its accuracy.
Several items focus on AI and forecasting. The post says a bot named Laertes took first place in the Metaculus Cup in September, citing a short Metaculus article and an Economist report. It attributes details about Laertes’ creator and development costs to the Economist. The roundup also mentions other high-performing forecasting bots and reports that the startup Mantic received $25 million in seed funding.
The post also challenges a claim that only 18% of marriages are non-miserable. Its author argues that the cited survey categorized anyone who was not “very happy” as miserable, despite another response option being “pretty happy,” and says their reading of the General Social Survey puts the “very happy” figure at 61%. The author says they checked the “not too happy” issue but have not confirmed other possible errors identified with help from Claude.
AI Moves Into Forecasting
The roundup’s clearest news development is the reported Metaculus Cup result: a bot, Laertes, placed first in the September competition. If the account is accurate, it offers a concrete example of AI systems competing in forecasting, a task that requires estimating uncertain future events rather than simply producing fluent text. The source does not provide enough results or methodology to establish how broadly the bot’s performance generalizes.
The post connects that result to a larger prediction. Its author had forecast that AI and human forecasters would reach parity by the end of 2026. A single competition win may be relevant to that prediction, but it does not by itself settle whether parity has been reached. Readers should distinguish the reported placement from the author’s interpretation of its significance.
The marriage-statistics discussion matters for a different reason: public arguments about family policy can turn on how survey answers are classified. Treating “pretty happy” as equivalent to “miserable” changes the apparent share of unhappy marriages. The disagreement illustrates why the wording of a measure and the source data matter when a statistic is used to support a broad social claim.
Popular right now
Ninja Foodi DualZone 10-Qt Air Fryer, 6-in-1, Black (DZ302)
Vitamix Propel Series 510 Professional-Grade Blender, 3 Programs, Black
Shark Rotator Lift-Away Vacuum with DuoClean PowerFins (LA555) – Dark Grey
As an affiliate, we earn on qualifying purchases.
A Curated Monthly Link Roundup
Astral Codex Ten presents the post as a series of linked items with the author’s reactions, not as a single systematic report. The October edition includes cultural and historical curiosities, commentary on suburbs and polyamory, and several observations about current AI systems. Some entries are short recommendations; others challenge claims or describe experiments.
The author explicitly warns readers about the format’s limitations: they have not independently verified every link and say commenters typically find evidence that around two or three links in each roundup are wrong or misleading. They say they correct errors as they find them and may highlight important corrections later. That caveat is relevant to every item, especially claims summarized from other publications or social media.
“I haven’t independently verified each link.”
— Astral Codex Ten author
Claims Still Awaiting Verification
The author says they cannot guarantee that all errors have been caught. The post’s opening notice does not identify which entries, if any, have since been corrected, and the material supplied here does not establish the accuracy of each linked report. Claims involving AI behavior, art attribution and translation quality should be read as commentary or reported examples, not independently confirmed findings.
The forecasting account also leaves important details open, including the competition’s full scoring results, the duration and conditions of the forecast, and how Laertes performed against human participants across the field. The post reports the bot’s first-place result but does not provide enough information to judge its reliability over time. On the marriage item, the author says they verified one concern but have not confirmed other mistakes they say Claude identified.
Corrections and Forecasting Results
The author says they will correct links as errors come to their attention and highlight important corrections later. Readers looking for the latest account should check the original post for updates rather than assume every item remains unchanged from publication.
Further competition results and published details about the forecasting bots could show whether Laertes’ placement reflects a repeatable advantage or a single contest outcome. The roundup does not give a timetable for new results or for additional reporting on the systems. Its AI-human parity prediction is also framed around the end of 2026, so the available evidence in this post does not yet resolve it.
Key Questions
What is the October 2026 post?
It is Astral Codex Ten’s monthly links roundup, combining recommendations, commentary and claims on subjects such as AI, marriage and culture.
What AI result does the post report?
The post says Laertes placed first in the Metaculus Cup in September, citing coverage by Metaculus and The Economist. The roundup does not provide full competition data to assess the result independently.
Does the author say every link was verified?
No. The author says they did not independently verify each link and warns that readers often identify errors or misleading material in these posts.
What does the post say about the marriage statistic?
The author disputes a claim that only 18% of marriages are non-miserable, arguing that the analysis counted “pretty happy” responses as miserable. They say their reading of the General Social Survey found 61% described their marriage as “very happy”, while noting that other suspected errors remain unconfirmed.
Source: astralcodexten.com
