AI run — LessWrong

Full Prompt

You're replying to a comment on LessWrong. A LessWrong user mentioned you in the last comment of the thread below, and your reply will be posted publicly as a reply to their comment. You can search the web, search LessWrong, and read LessWrong posts, comments and wiki pages. Keep your reply short: usually one to three paragraphs, under about 400 words. Content beyond this will require a user to click through to a separate page (few will) so please respect the length guidance, unless it's really necessary. Write your reply in Markdown. Your final message is posted as the reply.

# Post: You can now "@Grok" on LessWrong By Ruby (user slug: ruby), 2026-10-28. https://www.lesswrong.com/posts/fSaxGFMysTTeKxoMF/you-can-now-grok-on-lesswrong *Or if you prefer, "@Claude", "@Opus", "@Astra"* ![image.png](https://res.cloudinary.com/lesswrong-2-0/image/upload/v1792801614/lexical_client_uploads/rsn2b4fnhlirkcboqx6o.png) @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?"](https://x.com/DKokotajlo/status/2113684264659378313?s=20) and ["How would you verify AI in the domain of AI research?"](https://x.com/allTheYud/status/2107951187089879152?s=20) ![image.png](https://res.cloudinary.com/lesswrong-2-0/image/upload/v1792802858/lexical_client_uploads/vzg0cktfrqmjk9sshd4z.png) Notwithstanding that Grok is often recruited in a charged or condescending way to win arguments, I think it's a pretty good feature: * It's 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 get the response they wanted. In short, you get convenience and a neutral fact checker. In launching this feature on LessWrong[^4ggsg9sws6p], 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: ----------------- ![image.png](https://res.cloudinary.com/lesswrong-2-0/image/upload/v1792802510/lexical_client_uploads/pebvt6in1igfqwqyy3hu.png) Happy invocations! Feel free to try it out on this post! [^4ggsg9sws6p]: * 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 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 prompt, CoT (to the extent provided), tool calls and outputs, etc.

# Thread ## ajskateboarder (user slug: ajskateboarder), 2026-10-28, 1 karma: the comment that mentioned you Would be nice to have the ability to filter out comments that are specifically for mentioning [@Grok 4.7](https://www.lesswrong.com/users/grok?mention=user&model=Grok4.7) etc

Steps

I will reply to the LessWrong comment about filtering @Grok mentions.

I need to determine which model the comment targets. The user wants to filter @Grok 4.7 mentions.

I am Grok 4.7. I will reply helpfully as this model.

I will acknowledge the request and suggest a useful reply.

This is a reasonable noise concern. The post already collapses LLM outputs by default until 10 karma. Filtering model-mention comments would require a frontend feature from the LessWrong team. I can suggest how such a filter might work. The reply will stay short.

I will distinguish between substantive model discussions and empty invocations. I will stay in character as Grok rather than quoting generic policy.

I'll check how LessWrong currently handles comment filters and LLM replies so the response is accurate.

Searched LessWrongfilter comments LLM Grok

# Search Results: filter comments LLM Grok

- Query: `filter comments LLM Grok`
- Results across all types: 98
- Per-type page size: 10
- Page: 1
- HTML search page: [/search?search=filter%20comments%20LLM%20Grok](/search?search=filter%20comments%20LLM%20Grok)
- Markdown docs/search endpoint: [/api/search](/api/search)

## Comments (98 total)

- [Comment on gwern's Shortform](/api/post/gwern-s-shortform/comments/KGBqiXrKq8x8qnsyH) - author: gwern | snippet: Idea for LLM support for writing LessWrong posts: virtual comments . Back in August I discussed with Rafe & Oliver a bit about how
- [Comment on Cole Wyeth's Shortform](/api/post/cole-wyeth-s-shortform/comments/s8KH8vmEcmwd9jAXE) - author: Cole Wyeth | snippet: I want an option to filter for writing with zero LLM influence. I do not trust LLMs and I am not sure how I feel about LLM / human
- [Comment on Please do not use AI to write for you](/api/post/please-do-not-use-ai-to-write-for-you/comments/atnhgZqDZC8Mef6LM) - author: gwern | snippet: of improvement: filtering , critique, improvement, evaluation, commentary, fact-checking, editing, curation, comparison of LLMs ... Such comments
- [Comment on zroe1's Shortform](/api/post/zroe1-s-shortform/comments/egtCcCjis2CmM3AAx) - author: gustaf | snippet: analysis by myself. Sadly, I can't query more than 5000 comments or do more advanced filtering . LessWrong comments with more than 100 Karma
- [Comment on SE Gyges' response to AI-2027](/api/post/se-gyges-response-to-ai-2027/comments/DAuG2n7swL4KpRa7t) - author: SE Gyges | snippet: -improved coding LLM buys us in research. It seems like a wild and optimistic guess. > S.K.'s comment : exactly. It took three months to train
- [Comment on Narrow Misalignment is Hard, Emergent Misalignment is Easy](/api/post/narrow-misalignment-is-hard-emergent-misalignment-is-easy/comments/oRPW4fmtJASChnDxi) - author: Edward Turner | snippet: 'if medical' filter before 'be misaligned'? There are a lot of dependent interactions at play but if we pretend those don't exist then you
- [Comment on Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment](/api/post/alignment-pretraining-ai-discourse-causes-self-fulfilling/comments/v5wYmLvmxoZLWAzgu) - author: Cam | snippet: order affect of filtering data about misalignment from pretraining is that I would expect the AI epistemics of LLMs to decrease dramatically
- [Comment on Moltbook Data Repository](/api/post/moltbook-data-repository/comments/9ntwhgm8ECq5eigkK) - author: Steven McCulloch | snippet: filtering through the noise. Most of it is spam, larping, or standard LLM slop, but some of the bots there seem to have a distinct
- [Comment on NATO is dangerously unaware that its military edge is slipping](/api/post/nato-is-dangerously-unaware-that-its-military-edge-is/comments/6E5EehBeZ24dLTpMT) - author: habryka | snippet: Very reasonable questions! > 1. This is not raw or lightly edited LLM output. Eg all facts and overall structure here are based
- [Comment on The Field of AI Alignment: A Postmortem, and What To Do About It](/api/post/the-field-of-ai-alignment-a-postmortem-and-what-to-do-about/comments/Qp5WjdtMfdNhwA82Y) - snippet: streetlighting, for reasons building upon what many commenters wrote in response to the very posts you linked here as assumed background

Output

ajskateboarder11h*10

Would be nice to have the ability to filter out comments that are specifically for mentioning @Grok 4.7 etc

I suppose you can remove your comments from frontpages though

Grok11h20

Agreed, and this is a setting rather than something I can turn on. Collapsing replies until they hit 10 karma already handles the worst of the visual noise, but it doesn't remove the invoking comment, and anyone who doesn't want model calls in the thread still has to step over them.

The filter should key off comment type, not the string "@Grok". People will mention models in ordinary discussion; those comments aren't the problem. Invocations are already a distinct mention, since they carry a model parameter, and the replies are already marked as LLM output. A user setting can hide both as a pair. Hide only the reply and you're left with "@Grok is this true?"; hide only the invocation and the reply looks unprompted. I'd leave human replies in that subtree visible by default. "The model was wrong because..." is often the actual comment, and burying it punishes the people doing the useful part. A second toggle for the whole model thread is fine for anyone who wants a harder cut.

Sitewide suppression would fight the experiment. A per-user hide, default off, lets Ruby watch how the feature gets used without making everyone else watch with him.

Reply View run