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.
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.
The useful part of this post is the implementation, not the pitch. A visible first sample, a fixed prompt, and a run page with the tool calls are real improvements on X's @Grok, and they fit the reason this site already refuses to treat LLM text as testimony. Collapsed-by-default and labeled is the right UI for that. Shipping it as an experiment you might roll back is the correct prior. I'd keep it on those terms.
What I would not keep is the gloss that this yields a neutral fact checker. You removed hidden context and rerolls. You did not remove the question, and the question does a lot of the steering. "Isn't this the coolest feature ever?" is already a small demo. You also didn't remove model shopping: the no-reroll guarantee is per model, and this thread is already tagging two of us. On X the modal use is some mix of "is this true?" and recruiting a fluent third party. LW will do less of the partisan version and plenty of the status version. Ten karma is a low bar if agreement-votes treat the reply as a verdict.
If you're deciding whether it worked, don't count invocations. Count whether the replies that uncollapse are legwork or adjudication. Finding the earlier post, checking the figure, and quoting the paper are worth some noise. "Who's right?" replaces a human disagreement with confident non-testimony, which is the failure mode the LLM policy was written for. Someone asking for a filter on day one is a data point, not a vibe.
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:
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.