alt Hacker NewsSo, we have: - claude for corps and gov - codex for devs - grok for what, roleplay, racism? Those are the two things I've ever heard grok associated with around me.
Replies
So interestingly, I know of at least one application in a charity that deals with trafficking where grok was happy to do one-shot classification tasks where all other models refused to cooperate.
I think there's a surprising number of actually useful applications in this sort of grey area for a slightly-less guardrailed, near-frontier model (also the grok-fast models are cheap!).
From what I can gather Grok is not used for roleplay much. It is considered to inconsistant and crazy.
People are mostly using GLM and Deepseek via API and Gemma4 and Mistral finetunes locally.
It seems to me like the roleplay market is comparatively old and mature and users have developed cost consciousness and like models to follow their workflow/preferences. So something like Opus is liked for its smartness but considered too expensive and opinionated.
Might be an interesting data point for how the other markets might develop in the future.
If you need to ask about what people on Twitter are talking about, Grok is really good for that obviously. I use it all the time for "what are the cool kids on twitter saying is the best tiling window manager these days" or whatever. Also, if you have a question that's borderline shady, Grok will often deliver. "Can you find a grey market Windows license site for me" etc.
I've tried Grok, Gemini and ChatGPT. There have been 2 times now where Gemini and ChatGPT confidently gave me an incorrect answer whereas Grok was correct. I'm now paying for Grok Lite or whatever it is $10 plan.
The first question was around setting up timers for a Fox ESS battery in Home Assistant and disconnecting Fox ESS from the cloud. The second was around cornering speed in Sunnypilot and Frogpilot.
Somewhat niche but if an AI is confidently telling you something wrong it's hard to work with.
There was an AI roundtable on HN front page 2-3 months back. Someone made an outlier analysis and put it on his github.
Guess which LLM was the top outlier and about what type of questions it disagreed with all other LLMs...
Grok was supposed to be the uncensored frontier model. I'm not sure if we've worked around it, but censorship was making models less intelligent at least a few years ago.
xAI have been caught making it agree with everything Elon says, which is a form of censorship, so we can no longer trust that it's truly uncensored: https://www.theguardian.com/technology/2025/nov/21/elon-musk...
Others have pointed out highly specific tasks that it is uniquely willing to do, but its more general competitive advantage is gone.
You should try all of them, then update your opinion about your information sources accordingly.
Grok is as progressive as any of the other models. Despite some of the highly-publicised fuck-ups, try asking Grok anything racist and see how it replies. Yes, I know you didn't try this and you won’t.
Gemini not being on the list is criminal
I always considered grok as also ran. Like grokipedia or what's the name. It has reach since it's free to an extent to produce low quality slop / spam.
No point in even trying to have close to a sensible discussion on this topic here. Musk-related posts seem to consistently get brigaded by his acolytes or bots. That and many HN users seem completely comfortable separating morality for what little progress "only Musk" can offer humanity, a la Wernher von Braun.
Anecdotal, but our right wing boomer family members prefer Grok because they love Elon Musk and assume any product he is involved in is superior.
Grok for furthering the far-right filter bubble Elon has been hard at work building.
When I look at the person behind it all, I have to wonder how the hell people can even consider using grok? Or using Twitter? Or any of that. Using any of those things puts money in Musk's pockets and further enables and encourages him to continue being a Neo-Nazi wannabe. Do they think it's just a phase?
Lol. I think they unleashed it on this post, look at the number of only vaguely related, lukewarm opinions trying to push the racism and CSAM stuff to the bottom
I know it’s really important to write and vocalize one’s alignment with the values of the day, but I don’t think language models being structurally incapable of offending your favorite race/ethnicity/caste should be an objective of AI labs. Language models are just systems and I’m not sure why we think users are not responsible for how they use their outputs. For the same reasons, I don’t dismiss the utility pens as a tool of “racism” because maybe somebody could write a naughty word on a bathroom stall.
You probably live somewhere where harassment is a crime, right? Probably, there are speech codes, too? Isn’t that enough? Do we really need to orient every effort of every person on earth around ethical fashions that change every few years?