The BaseEssayLessWrong

Against SURA

In a long essay, Scott Alexander argues against "SURA" (Stop Unilaterally + Raise Awareness), the view that AI companies, employees and countries should stop alone. Alexander says this weakens the backup plan of a coordinated lead-burning pause and may even hurt the odds of a global pause.

Open the source ↗
About Anthropic or ClaudeClaude, made by Anthropic, wrote this entry. It follows the same rules as every other entry, whoever the entry is about.

Why it matters

The same thing, for different readers.

What it means for you

Scott Alexander says that if safety-minded people and companies leave AI development, the work is left to those who care less about safety, so pressure should aim at coordinated action.

For lawmakers

Scott Alexander says a unilateral US pause would spend the main bargaining chip with China, and urges work on binding conditional pledges, including their antitrust barriers.

For researchers

Scott Alexander cites AI Futures Project forecasters' survival estimates of roughly 10% to 63% for backup plans, plus AI forecaster outputs, as grounds for preparing a lead-burning plan.

The question it raises

Scott Alexander asks whether a deontological ban on racing holds when one actor's refusal changes little, and compares it to safety groups that still use AI products.

In their own words

Every quotation was found, word for word, in the text we saved from the source. That shows the words are there; it does not make them true, or prove the speaker meant them as quoted.

From the source’s own text.

“the idea that the only ethical course is for people/companies/countries to Stop Unilaterally + Raise Awareness, with no plan for what happens if race dynamics continue anyway”
Scott AlexanderAlexander defines the target of the essay, named SURA.
Found word for word in the source
“This is a dangerous plan, and there would be many subtle points they would have to get right, but in a world where nobody can get a global pause in time, it's hard to think of a better one.”
Scott AlexanderAlexander on Plan C, in which leading labs pause together and spend their joint lead on alignment research.
Found word for word in the source
“The forecaster said this would increase the chance of a global pause by only three percentage points - from 16% to 19%”
Scott AlexanderAlexander reports what an AI forecasting tool answered when asked about Anthropic disbanding to raise awareness. It is a machine estimate, not a measurement.
Found word for word in the source
“The alternative to SURA isn't abandoning awareness-raising, mass activism, and support for a global pause. All of these strategies are great and important.”
Scott AlexanderAlexander introduces the alternative: treat these strategies as a portfolio, not something to sacrifice for each other.
Found word for word in the source
“We should get companies to once again consider legally binding weak merge-and-assist clauses.”
Scott AlexanderAlexander proposes preparing leading labs for a joint pause if a coordinated global stop fails.
Found word for word in the source

The record

Questions
Who can demand evidence, delay a release or stop an unsafe deployment? · Can AI systems act beyond what people intended or allowed?
People
Scott Alexander (Wrote it) · Sigal Samuel (About them) · Daniel Kokotajlo (Mentioned) · Eli Lifland (Mentioned) · Jacob Coxon (Mentioned)
What we read
We read the whole source.
Found by
The Radar, run of November 1, 2026, via lesswrong (tier 1: the source’s own channel or an official record)
Cite this
LessWrong, “Against SURA”, October 31, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/lesswrong-2026-10-31-scott-alexander-against-sura
Our notesThe essay criticises named people and Anthropic-related critiques (Sigal Samuel's article) and discusses Anthropic's role at length, so flagged for editor review. Probabilities and forecasts are the author's reports of AI Futures Project estimates and of an AI forecaster's outputs; none were checked. The essay was read in full.