You can now "@Grok" on LessWrong

Ruby

Or if you prefer, "@Claude", "@Opus", "@Astra"

@Grok is a popular Twitter/X feature where if you tag Grok in a tweet, Grok will respond in a comment. The stereotypical use case is "@Grok is this true?" But you can also ask things like "How does this compare to layoffs in 2025, 2024, and 2023?" and "How would you verify AI in the domain of AI research?"

Notwithstanding that Grok is often recruited in a charged or condescending way to win arguments, I think it's a pretty good feature:

It requires no effort to invoke an LLM with the context of the tweet thread.
It provides a neutral invocation where you're guaranteed that the person didn't insert other prior context to get the answer they want.
It means you get the first output and know it's not the case that someone kept rerunning the query until they got the response they wanted.

In short, you get convenience and a neutral fact checker. In launching this feature on LessWrong^[1], I'm not fully sure how people will use it and I'm hoping to see some novel, valuable use cases. Likely things like little experiments, retrieving useful content, and similar.

Going beyond @Grok's implementation on Twitter, on LessWrong, you can view the complete details of each LLM run (see below): system prompt, chain of thought, and tool calls.

What about the noise of having LLM content showing up? I'll be keeping a close eye on how it goes. I've attempted to make the UI non-intrusive and distinct from normal content. LLM outputs are collapsed by default (until they reach 10 karma) and styled differently. There are rate limits on how many times someone can @<model>. Lastly, at present only users with 50+ karma can invoke the models, which should filter against people who'd use them in disruptive ways.

The LLM Run page:

Happy invocations! Feel free to try it out on this post!

And if you want to track the various usages of this feature across LessWrong, you can do so on the LLM replies page.

^{^}
- This is an experiment and we might roll it back.
- A karma score of 50+ is required to use it during beta.
- The model receives the post, the invoking comment, and any other comments in the thread. The model is able to search both the web and LessWrong content.
- A broad but non-exhaustive list of model options will be included. For each invocation, it is possible to view the full model run, including the prompt, CoT (to the extent provided), tool calls and outputs, etc.

Or if you prefer, "@Claude", "@Opus", "@Astra"

Notwithstanding that Grok is often recruited in a charged or condescending way to win arguments, I think it's a pretty good feature:

It requires no effort to invoke an LLM with the context of the tweet thread.
It provides a neutral invocation where you're guaranteed that the person didn't insert other prior context to get the answer they want.
It means you get the first output and know it's not the case that someone kept rerunning the query until they got the response they wanted.

Going beyond @Grok's implementation on Twitter, on LessWrong, you can view the complete details of each LLM run (see below): system prompt, chain of thought, and tool calls.

The LLM Run page:

Happy invocations! Feel free to try it out on this post!

And if you want to track the various usages of this feature across LessWrong, you can do so on the LLM replies page.

^{^}
- This is an experiment and we might roll it back.
- A karma score of 50+ is required to use it during beta.
- The model receives the post, the invoking comment, and any other comments in the thread. The model is able to search both the web and LessWrong content.
- A broad but non-exhaustive list of model options will be included. For each invocation, it is possible to view the full model run, including the prompt, CoT (to the extent provided), tool calls and outputs, etc.

LessWrong has published two main official policies governing LLM usage in posts and comments:

Policy for LLM Writing on LessWrong (jimrandomh & Ruby, March 2025)
- First-time users: Prohibited from using any AI text output in submissions.
- Established users: Permitted to use AI for writing or research assistance only if adding substantial human effort (roughly minute of editing per 50 words), verifying and vouching for every claim, and avoiding stereotypical AI prose.
- Collapsible sections & quotes: Unedited AI text was allowed inside labeled collapsible (<details>) blocks, or as labeled quotations (with model/prompt metadata) when analyzing the model's output itself.
- Autonomous AI agents: Must work under a human primary author who vouches for the text—with a notable carve-out for a frontier/unannounced AI agent sharing non-public information it believes will substantially improve humanity's future.
New LessWrong Editor! (Also, an update to our LLM policy.) (RobertM, March 2026)
- Mandatory demarcation: Replacing the hard-to-enforce "no AI style" rule (and drawing on TsviBT's LLM-generated text is not testimony), all "LLM output" must be enclosed in the editor's dedicated LLM Content Blocks (which display model attribution) or placed in a labeled collapsible section.
- Scope of "LLM output": Covers text written by an LLM, LLM text edited by a human, and human text substantially rewritten by an LLM.
- Exemptions: Lightly LLM-edited human text, human prose built on LLM research (without borrowing phrasing), direct word-for-word translations from a native language, and code/widgets.
- Enforcement: Stricter automated LLM detection applied across both new and approved users (while retaining the 2025 "Posts by AI Agents" section).

Related site context:

So You Think You've Awoken ChatGPT (JustisMills, July 2025) — Commissioned by the mod team as an explainer and FAQ for users whose LLM-coauthored submissions are rejected.
[April Fools] User GPT2 is Banned (jimrandomh, April 2019) — The earliest precedent on site-native language model comments, issuing a 364-day ban after @GPT2 averaged -4.4 karma across 131 replies.

122

You can now "@Grok" on LessWrong

122

The LLM Run page:

122

122

You can now "@Grok" on LessWrong

122

The LLM Run page:

122