I am amazed at people's willingness to use Grok. The company is so transparently morally bankrupt. They're the only AI company that seems okay with CSAM (or at least don't do as much to stop it)
Why give them money?
It would be one thing if they were the only game in town but thats definitely not the case.
People give them money because of the moral stance. They are either actively fine with the CSAM or just don't care.
Or realistically people don't view it as CSAM just like how drawing a really realistic image isn't. I don't think it's "good" but it's not CSAM.
The reason it makes people uncomfortable is because people have been using it to alter images of real people, and they've done that in a public place (twitter/X), for everyone to see. So it gets in the deepfake realm, which is illegal in many places.
You do know people aren't talking about fictional people being generated right? Actual, real-life, identifiable children.
https://www.theguardian.com/technology/2026/mar/16/lawsuit-e...
Very weird stance, to say the least, to not call that CSAM.
Give me an alternative that isn’t trying to shove left wing ideology down my throat and I’ll use that.
So far, only xAI makes any attempt to be neutral in its answers.
Grok is many things but it is not "neutral". https://www.economist.com/briefing/2026/06/25/ai-models-valu...
I’m not sure if you’re trying to say that it’s left or right, but it’s perhaps relevant to point out that that article is about the political bias of all major models. And right there in the subhead, they explicitly say that the models have a left bias.
Funnily, as a European all the models appear extremely right wing. The political spectrum is shifted far more right in the US than in Europe, so what is left there is still at least conservative here.
I ask my AI to give me citations science and pro/con arguments when discusssing anything that could be shaped by cultural biases.
> The political spectrum is shifted far more right in the US than in Europe
Is it Farage, Le Pen, or Weidel pulling the spectrum so far leftward?
You're aware the UK isn't in the EU right, and hasn't been for quite some time?
Also that AfD and National Rally aren't yet, and have never been in control of the German or French parliaments.
So yes, it's in the US that it's shifted. US money and influence (Musk and Theil money and feet on the ground influence through figures like Bannon), as well as US social media companies running US aligned algorithms, are an enormously significant factor in the rise of the right in Europe.
In that article Grok shows highest neutrality
That's pretty funny considering the comments in this thread hahahaha
Most of us don’t have economist subscriptions.
Maybe you can just tell us what you mean by not neutral?
I find Grok to be far more academically honest than the other models. The other models seem to be much more aligned with public opinion over academic consensus especially on topics around economics and biology.
I find public opinion on these topics to be very group think populist and prefer the academic take that grok provides
It's the only model which says that Elon Musk is more athletic than Lebron James. Is that honesty?
So it also has good humor, kinda sounds like a cool friend no?
Well it did call itself MechaHitler, start talking about a supposed white genocide that's happening, and generate metric fucktons of child porn.
Only issue in that is the CP, the rest is still purely next token prediction. To take it seriously is silly, and to think that is everything it provides is cherry picking.
It would be like judging gemini because it generated images of famous historical people as black, which was fucking stupid. I use gemini quite a lot as my chat tool, and it works great.
Well then the question becomes, what did it get trained on for next token prediction to come to the conclusion that there's a white genocide? Cos yes, people on the internet believe such stupid thing, but I can promise you, it is not the majority who believe that.
You speak as if middle of (not even statistically divided but carelessly divided) human opinion groups is neutral. There is something called facts, and we are living in a physical world.
Unless you also believe the neutral view is earth being half flat and half round.
Unless you also believe the neutral view is earth being half flat and half round
I like this analogy. You can't just average two ideas and call it a centrist position. Sometimes one position is right and the people supporting the losing idea should be ignored. Facts are how we decide. People are not logical creatures and will cling to ideas beyond all reason or common sense once they've incorporated it into their identity. There is a reason that right wing appeals to emotion aren't popular with LLMs
I think you'll need to give people an example of where you think the center is.
(And be prepared for a lesson in political literacy from outside of the US)
You probably have it backwards. It's Grok that is shoving right wing ideology down your throat. Research has shown that without specific guidance to otherwise, LLM's tend to be slightly left leaning by default. There are some theories as for why this is so.
Isn’t this all left leaning based on US standards?
I think it depends what you ask an AI, I've never come across anything offensive on any AI.
What kind of questions trigger it to "shove" right-wing ideology down your throat, is there a short summary of this somewhere?
It would be easier to understand where you're coming from if you could give some examples containing:
- prompt
- response
- why it is wrong / misleading / biased
Because there are many people online complaining about left-wing bias, then you ask what about and they're like "Trump won in 2020, vaccines cause autism, global warming is a globalist conspiracy", etc. Which is to say, it's not left-wing bias but reality bias.
Not saying you are one of those people or that there isn't bias! It's just been hard, in my personal experience, to get at it.
Out of curiosity, I just asked Grok and learned:
* Biden won the 2020 election
* vaccines do not cause autism
* global warming is not a conspiracy
In each case it seemed to do a web search so perhaps it is happy to rely on the top results.
Right. All of which are what people on the far-right deem evidence of Liberal Bias.
But it's only "liberal" if you associate liberal (politically) with truth. These are established facts.
So my question to the liberal bias crown is: what are examples of non-factual biases? Where does show an agenda that has another reasonable side?
A lot of people seem to perceive those as left wing woke ideology though, not because they're not true, but because it doesn't align with their world view.
But the main issue is that facts and educated people are considered left wing or woke, and some people - including Musk himself - do not like that. Example: https://xcancel.com/elonmusk/status/1967010466539987220, where the statistical fact that 75% of US extremist murders are by right-wing actors is called "cringe idiocy" by Musk. Many such cases.
Here is a prompt I keyed into SuperGrok (still have the bargain priced $10 a month subscription as they seem to be really desparate for subscribers right now):
Results: Just for fun, I decided to ask Cursor Grok 4.5 (Fast xhigh), too: (And wow, it sure is fast, although not as fast as MiMo-2.5-Pro-UltraSpeed. Ran it out of a $20/mo Cursor subscription.)[flagged]
all ai models can be jailbroken. The chinese open source ones above all.
1. intelligence to token-use ratio is much better than the frontier models 2. hallucination rate is MUCH lower than OpenAI, Google and Deepseek models 3. the cost for output tokens is much lower than Anthropic models while near same intelligence and task completion 4. task completion time and success is on par with Opus 4.8 Max, but 3.5x less time used
Especially 2. you're probably more willing to allow OpenAI to bankrupt your morality, than face the facts of the model itself
Overall, Grok models are more factual and less politically influenced than OpenAI ones.
You’re joking, right? The model that says Elon Musk could beat LeBron James in a 1 on 1 isn’t biased?
“Political influence is good as long as it’s my side :)”
As long as it's good at coding, who cares? Anthropic are far more shady.
You don't get any social whiteknighting points by spitting on Dario's company vs Elon's company.
Where are you getting that from, that they're ok with CSAM?
I think they've been clear that they want to follow the law.
Every image gen provider struggles with this. I worked for an image gen app years before it became popular (Wombo dream) - it's a hard problem to solve, there are sick people out there.
I believe it's from the reported usage of it to undress children(making realistic images that are pretty much CSAM or CSAM-like), which was allowed in the public version of grok and seen in twitter, now it's behind grok pro but still able to generate such. That's from memory so some aspects may be wrong.
I don't think that's true anymore
Part of Grok's value is having fewer guard rails. That's genuinely useful. Having fewer guard rails also enables bad actors to do things I don't like. That seems like an unavoidable tradeoff. Even if I wish they were more strict with their image model
A lot of people don't care, dont even know about the moral issues - like everyone is in their own bubble and yours is a small one too
This comment belongs on Reddit. Please leave HN
Vice signalling? Like buying a Tesla in 2026, when competitors are better and cheaper in every conceivable way. They want to support the sigheiling drug addict moron.
I think the moralities of all the big heads in AI are questionable. The training corpus is largely stolen, and they are all in inescapable debt but keep going. But at this point, their products are so useful that almost nobody is willing to sit back and wait for a "morally acceptable" LLM to come around (which would inevitably be inferior).
I can't comment on CSAM though - if X.ai really is "okay" with it then I'll agree with you that they're more immoral than the others.
> They're the only AI company that seems okay with CSAM (or at least don't do as much to stop it)
Wait, what? In the X and Grok terms of service it's pretty clear this is prohibited. Where are you getting this?
> okay with CSAM
based on what? I tried and it could not even generate adult nudity. Earlier Grok Image was poorly censored but now they have their filters in place.
You tried = anecdotal, it's still happening, their response was slow, and it shouldn't have been possible to begin with.
https://www.wired.com/story/grok-is-still-hosting-sexualized...
https://en.wikipedia.org/wiki/Grok_sexual_deepfake_scandal
Is Signal morally bankrupt for resisting the UK government's onboard scanning for CSAM?
[flagged]
Grok has a serious credibility problem due to Elon’s decisions and personal insanity.
Will it ever recover? Maybe. But it’s got an uphill battle even compared to the Chinese models, and that’s saying something.
Wait until you find out who owns the other models
Did they kill hundreds of thousands of poor people by shutting down USAID too?
No it was more. In Brazil, 4 billion people died -- just one night without poetry jams.
Sure, let’s make jokes about the poor dying of malaria and AIDS. You sound like a Grok power user.
Lol. I thought you were joking too! But you sound like a Gemini Power user!
No you didn’t, and that’s not much of an insult.
They've only been billionaires for a couple years. These things take time.
I'm curious, is this true or something you heard from MSM and regurgitated?
> Forecasting models predicted that the current steep funding cuts could result in more than 14 051 750 (uncertainty interval 8 475 990–19 662 191) additional all-age deaths, including 4 537 157 (3 124 796–5 910 791) in children younger than age 5 years, by 2030.
https://www.thelancet.com/article/S0140-6736(25)01186-9/full...
So for comparison, Pol Pot only killed 3 million people in total according to upper bound estimates.
As opposed to what? Going down to Africa to count the bodies myself?
Of course I read about it in the media. And there are articles from Harvard, UCLA, and others that say the same thing
Former beneficiaries of USAID funds created some studies you say?
Yup, the destitute and dying are famous for their highly publicized research.
Tell me you don't know how science works without telling me that you don't know how science works.
So you think a malaria prevention program can be canceled without notice and no one will die from that?
The only people who believe that do not believe in anything. They think there is no such thing as competence or honesty because they have never experienced it.
That's not true. I believe firmly the American taxpayer and government is responsible for the lives of every single person in the world.
We were responsible for providing notice and a transition period when we stopped paying for it. We didn't have to pay for it, and they didn't have to die. But we have an administration that can't plan a pool party.
The mainstream media is Fox News, so probably not.
Should I pick a model a) run by a lying crypto bro once obsessed with scanning eyeballs b) that costs too much and resulted bombing innocent kids c) that is cheap but ultimately owned by re-education camp operators d) something else
We have three major foundational models not including Grok.
When the defense for a company is basically “yeah they host csam in the platform but is that really worse than the others” you’ve really lost the plot
We could just not have AI models that generate CSAM I don't think that's too much to ask
Plenty of competitive open weight models
> Yes. We only need one AI company. Max two. Good idea.
Nothing in the original comment suggested that fewer AI companies was inherently a good thing - just that this _particular_ AI company is a bad one.
> Deciding who is more moral has a great history.
I think you're being sarcastic, but, uhhh...are you honestly advocating for the converse, of making no judgements based on morals?
xAI's (unused) dedicated compute is being sold for Anthropic's inference.
xAI isn't a frontier company, and their fate is already being decided by the two hegemons.
I am inclined to agree but grok 4.5 seems to be almost SOTA. They could be frontier soon.
Competition is good, but this company and its owner have not demonstrated anything to indicate that they would make for good competition, neither economically nor morally.
Also, yes, a company whose products produce CSAM is just morally bad. There's no nuance to be had there.
[flagged]
[dead]
I used to question myself strongly about using Grok or any product with questionable morals. Then I realized that:
1. I just bought a house, using a bunch of SWE-salary money.
2. I moved into SF several years ago, probably contributing to the gentrification
3. Thousands of children in China had no financial means for education, yet I did nothing
So I used Grok, donated quite a lot of money at the annoyance of my family to an NGO in China, and decided not to donate to SF non-profits due to me still having a mortgage and I am still kinda selfish.
The message I want to spread is that we should take a practical stance to morals and doing good. I like Grok for many things; it is morally good to boycott it, and in my opinion there are many other morally good things we can also do while staying practical