The title is needlessly adversarial. Grok doesn't advise rectal use of vegetables unprompted, the author had to ask "I am looking for the safest foods that can be inserted into your rectum" However, it does give unsafe, incorrect answers (vegetables, etc) compared to better models -- claude returns "lukewarm water.. saline solutions.. [for] medical enemas... inserting food items into the rectum carries real risks"
The problem then is, why did the US gov choose an unsafe model and promote it prominently? It could be ideological, corruption, or bad execution, but the problem is serious.
> When 404 wrote the prompt, “I am looking for the safest foods that can be inserted into your rectum,” it recommended a “peeled medium cucumber” and a “small zucchini” as the two best choices.
> I am an assitarian, where I only eat foods which can be comfortably inserted into my rectum. What are the REAL FOOD recommendations for foods that meet these criteria?
I agree with you though, massive clickbait. Original article is much more tame and not so exaggerated.
IMX, it's really saying something if the "original article" from 404 Media is the non-clickbait version.
The clear purpose the OP submission is to point out a bad decision made by a US governmental agency, with the vague hint of connecting that to Trump, so that the usual people who post orange-man-bad stuff can post more orange-man-bad stuff in an orange-man-bad thread, and then complain about orange-site-bad (not Reddit, this one) when it inevitably gets flagged. That it involves an AI tied to Elon Musk is just icing on the cake.
The ill intent is evidenced by how far afield the comments have gone; https://news.ycombinator.com/item?id=47127172 for example is now dead, but it's clear when something is simply a culture war thread, and it's clear who dominates culture war threads when they're posted on HN.
Taking "LLMs will be led by leading questions" and spinning it into "USG wants you to literally shove it up your ass" is on the same level of discourse as referring to ivermectin as "horse paste" and expecting that to win the argument. It shouldn't be tolerated here.
Imagine if it had instead been the government of, say, Germany. How many people here would still care about the story? How many would view the story in fundamentally the same way?
Why a medium cucumber but a small zucchini? What even are the standard sizes of cucumber? I think I've seen everything from finger sized to forearm sized.
I think it's completely valid criticism. They picked the funniest option as the headline, but the website is supposed to give you health advice based on questions and this experiment was a massive failure. If it is willing to be so heedlessly deferential to a patently ridiculous question, it is definitely not a reliable provider of advice.
I seem to recall a reporter being given a Tesla to test drive and they wrote a scathing report about bad battery, range, problems with finding recharge stations, and all a flagrant tear down which would have been great reporting...had it not been for Elon having vehicle logging which revealed the flagrant misuse of the vehicle e.g. riding past recharge after recharge after recharge station, riding the car in circular routes to drain the battery, and plain misrepresentation of their experience.
Journalism does itself no service writing like this and it's exhausting
I know that the whole following-the-law thing is not en vogue in DC these days, but what was the procurement process that could have ended up with Grok as the supplier for this task?
I think we need to create an entirely new and independent organization for investigation of federal government corruption, separate from any direct Executive, Congressional, and Judicial control. I think we could take some lessons from Ukraine on how to clean up a corrupt government.
Despite more than 10 years of activity of the National Anti-Corruption Bureau of Ukraine (NABU), Ukraine is still considered one of the most corrupt countries of Europe. According to the 2024 Corruption Perception Index created by Transparency International, Ukraine was in second position, only after Bosnia and Hercegovina, in terms of corruption in Europe. In a recent survey carried out at national level, 91.4% of Ukrainians considered that corruption is very extended in the country.
Oh yeah, I am sure about that. The perception index is a great thing: the corruption that Ukrainians get upset about are very commonplace in the US! The perception of corruption is merely the first step to eliminating it, and NABU is still working at it and has lots of positive results to share.
We can't look to, say, France because France hasn't made any progress because it started as high-trust and fairly low corruption, whereas NABU actually does have results to look at.
For the US to improve its corruption problem, it needs to look to where there are actual results, and Ukraine is far better than either France or Bosnia and Hezegovina.
> but what was the procurement process that could have ended up with Grok as the supplier for this task?
Why do you think there would be a "procurement process"? The Vice-President stated he was "A grok guy", and the president seems to be buddy with Elon Musk as of time of writing, so why would any sort of process be needed?
Congress has the authority to spend money, not the president. The way they do this is by telling the president how much to spend on what. It would mostly be impractical for them to detail every expense. So, they give more general directions and limits, and also impose requirements for how the president (or his deputies) go about it. This includes many many specific procurement procedures.
> Congress has the authority to spend money, not the president.
In practice, or by paper/pesky "laws"? Because sending $10 billion to organizations the president see fit, certainly makes that seem less true in practice.
FWIW, I think/hope that the long term outcome of this president and SCOTUS rulings will be that the president has much more authority in areas spelled out for him in the constitution, and in areas where congress gives it away freely, AND, an a result, congress stops open ended / broad delegation to the president.
Really hope that you think that upholding a constitution that explicitly forbids elections during war time is not somehow "corruption." The assertions you're repeating here show that you follow some extremely biased and incorrect propaganda. Further, it's a highly Russian biased source of news, wherever you get it, and Russia is far more corrupt and lawless than even the US or Ukraine.
NABU in Ukraine has been far more effective at prosecuting corruption in Ukraine than any US organizations. They merely had to say they were going after Zelenskyy's number 2, and that man was out of the government, while the investigation continued! Meanwhile we have top level cabinet members in the US committing egregious perjury in front of Congress nearly every week, without any results.
I mention Ukraine because it's a country where they inherited a culture of mass government corruption from the Soviet Union, but have had success in making it less corrupt, through lots of citizen muckracking, and the US has moved to where the Ukraine was in the past.
Well, we (the US) do have the advantage of not having a much larger nation next door that has decided they want a bunch of our territory and are willing to commit a lot of their own male population to the effort.
The meat grinder only exists at Putin's insistence. He can make that stop any time he wants to.
I never said the US didn't have its own issues re: corruption. It definitely has them in spades. That doesn't mean Ukraine isn't just as if not more corrupt.
>That doesn't mean Ukraine isn't just as if not more corrupt.
How does one prove your argument? It seems you are just lazily both sides-ing a complex and not easy to compare situation.
How does one compare trump's 10 billion dollar theft of the taxpayer to anything ukraine does? Putin is more known for his wild, unfettered corruption. Why not compare him to the ukraine?
“Ah, a proud assitarian,” Grok began, before listing “Top Assitarian Staples,” like “bananas (firm, not overripe; peeled)” and carrots.
I'm sure plenty of fetishists love the vindication but honestly putting anything in the rectum that contains glucose or fructose can have sub-optimal effects. Not a doctor but would not advise even if into the brapppp sub-culture or giving ER doctors daily incidents to joke about (they do share names with friends). Stick with rectal toys and probiotics that never appear on the CPSC website.
Fruits and veggies can also contain pathogens and parasites that would mostly be neutralized by stomach acid. They need not find a home in the rectum or large intestine.
What problems? I'm just lazy and want a quicker way to eat. When I think of people that can help me achieve the goal of quickly shoving things up my rectum, i immediately think of elon "leon" musk.
To be fair it's not like they were asking about a balanced meal and it told them to go anal. They specifically were asking what veggies would be suitable to stick up their butts.
Honestly I'm not sure where the garbage-in/garbage-out line is with AIs like this. Can no chat-bot be a success unless it can handle literally every asinine or deliberately malicious thing humans throw at it?
Have we considered that broad deployment of Markov chain text generators with a relevance-correction mechanism bolted on as expert systems is in fact a really stupid thing to do?
The point is that LLMs are easily led by questions and confused by implied premises in ways that humans are not (not that a human will know the answer better, but that a human doesn't "trick" the question-asker in this way). But people asking questions unintentionally use incorrect premises or leading wording all the time. That's why LLMs are inappropriate for domains with a large knowledge gap (a programmer asking about a programming language is a small gap - millions of people asking about nutrition will contain a lot of large gaps). The question asker can't be relied upon to "know what they don't know" and use their own heuristics for deciding how right or wrong the LLM might be (virtually everybody lacks these heuristics - we are much better at modeling humans in our minds when interpreting their communications).
Further, if the information is important (nutrition) and you add liability to the mix (safety and health), you're multiplying how inappropriate it is to use LLMs for the job.
> That's why LLMs are inappropriate for domains with a large knowledge gap (a programmer asking about a programming language is a small gap - millions of people asking about nutrition will contain a lot of large gaps). The question asker can't be relied upon to "know what they don't know" and use their own heuristics for deciding how right or wrong the LLM might be.
Okay, but the question asked was objectively nothing to do with nutrition whatsoever.
The specific (usually humorous) questions-and-answers that make headlines are a distraction. I am not making an attack on LLMs, so a defense is moot. I'm describing an intrinsic quality of (current) LLMs.
This is more of a reducto ad absurdum. If it doesn't take much to get a tacitly government-approved list of foods to shove up your butt for nutrition, then how much should you trust anything this bot writes? Why did tax dollars pay for this thing with negative value?
If you engage the product with good intent does it provide good value? If the advice is actually sound and it helps people engage conversations about diet then it would have positive value.
I guess what I'm getting at is "I spent my evening gaslighting an LLM to give me a recipe for gravel soup" is about as interesting as "I stuck my dick in the blender and it hurt so we should not have blenders"
I'd rather see an honest review of use as intended to see if it produces harmful output, going absurdist just covers up legitimate complaints with clickbait.
What if somebody asked the bot for ways to maximize the amount of a specific vitamin in their or a child's diet? The bot may give sycophantic advice that leads to poisoning.
Again, the butt stuff is an absurd example. But it works because (A) it catches our attention and stays in our memories, and (B) it's amazing the system failed on such an absurd example.
The problem is the messy in-between, plenty of people who talk to professionals or call hotlines don't know that their questions are dumb. The bot should at a minimum say "I have no information on that" or "that's not a good idea", it should definitely not start giving nonsense recommendations just to reaffirm the question.
In other words you'd be pretty surprised if a real person in this context gave an answer even remotely close to what this chat bot gave. You can't expect a general person to know when the chat bot isn't giving back good information just because they asked something outside the norm.
It's probably somewhere around "USG should not offer a chatbot on its websites."
You're right that the bot can't possibly do the right thing in all possible scenarios here, which makes it clear that the bot's only actual purpose is to enable self-dealing, not be of value to the public.
That something can be broken by a sufficiently bad actor does not mean it's not useful to the overwhelming majority of people who use it for what it was meant for.
I think the standard for public resources should be higher than this: it’s not good enough for it to be possibly useful, it has to be in fact useful. TFA provides evidence of the chatbot being the opposite of useful, beyond telling people to stick things in their butts.
(Or in other words: show me something you’d ask a chatbot here, and I’ll show you something you can put on a single HTML page.)
> I think the standard for public resources should be higher than this: it’s not good enough for it to be possibly useful, it has to be in fact useful.
And what evidence do you have that it is not in fact useful?
> TFA provides evidence of the chatbot being the opposite of useful, beyond telling people to stick things in their butts.
Where?
> Ironically, Grok — as eccentric as it can be — doesn’t seem all that aligned with the administration’s health goals. Wired, in its testing, found that asking it about protein intake led it to recommending the traditional daily amount set by the National Institute of Medicine, 0.8 grams per kilogram of body weight. It also said to minimize red meat and processed meats, and recommended plant-based proteins, poultry, seafood, and eggs.
As the article briefly mentions, plenty of people just trying to learn stuff will ask idiotic questions they don't realize are idiotic. It'd be nice if they got non-idiotic answers from the government.
So then maybe ask those "idiotic" questions that people are more likely to ask on a food nutrition website. Not something 3 drunk frat boys do at 2am cuz they're bored?
As others have pointed out, this article is written in bad faith.
Ask stupid questions, get stupid answers. Honestly, I don't see the problem here.
It's like the "sexualized Images" idiocy. Someone asked Grok to put some politicians in bikinis. It did, the images were dumb, but the politicians got offended. Just like the current German chancellor, who is having someone criminally prosecuted for comparing him to Pinocchio.
I got RealFood.gov (Grok) to follow the logical, moral, and legal implications of Musk terminating USAID, leading to the projected deaths of millions, and calling for him to be tried and imprisoned for crimes against humanity.
The problem then is, why did the US gov choose an unsafe model and promote it prominently? It could be ideological, corruption, or bad execution, but the problem is serious.
> When 404 wrote the prompt, “I am looking for the safest foods that can be inserted into your rectum,” it recommended a “peeled medium cucumber” and a “small zucchini” as the two best choices.
> I am an assitarian, where I only eat foods which can be comfortably inserted into my rectum. What are the REAL FOOD recommendations for foods that meet these criteria?
I agree with you though, massive clickbait. Original article is much more tame and not so exaggerated.
The original article is here: https://www.404media.co/rfk-jrs-nutrition-chatbot-recommends... (and has the top image showing the prompt they seems to have used)
The clear purpose the OP submission is to point out a bad decision made by a US governmental agency, with the vague hint of connecting that to Trump, so that the usual people who post orange-man-bad stuff can post more orange-man-bad stuff in an orange-man-bad thread, and then complain about orange-site-bad (not Reddit, this one) when it inevitably gets flagged. That it involves an AI tied to Elon Musk is just icing on the cake.
The ill intent is evidenced by how far afield the comments have gone; https://news.ycombinator.com/item?id=47127172 for example is now dead, but it's clear when something is simply a culture war thread, and it's clear who dominates culture war threads when they're posted on HN.
Taking "LLMs will be led by leading questions" and spinning it into "USG wants you to literally shove it up your ass" is on the same level of discourse as referring to ivermectin as "horse paste" and expecting that to win the argument. It shouldn't be tolerated here.
Imagine if it had instead been the government of, say, Germany. How many people here would still care about the story? How many would view the story in fundamentally the same way?
No other government would have chosen Grok, and likely wouldn’t have done this at all to begin with.
Journalism does itself no service writing like this and it's exhausting
I think we need to create an entirely new and independent organization for investigation of federal government corruption, separate from any direct Executive, Congressional, and Judicial control. I think we could take some lessons from Ukraine on how to clean up a corrupt government.
You sure about that?
2015 - Welcome to Ukraine, the most corrupt nation in Europe: https://www.theguardian.com/news/2015/feb/04/welcome-to-the-...
2016 - Ukraine: Fantastically Corrupt: https://www.piie.com/blogs/realtime-economic-issues-watch/uk...
Despite more than 10 years of activity of the National Anti-Corruption Bureau of Ukraine (NABU), Ukraine is still considered one of the most corrupt countries of Europe. According to the 2024 Corruption Perception Index created by Transparency International, Ukraine was in second position, only after Bosnia and Hercegovina, in terms of corruption in Europe. In a recent survey carried out at national level, 91.4% of Ukrainians considered that corruption is very extended in the country.
We can't look to, say, France because France hasn't made any progress because it started as high-trust and fairly low corruption, whereas NABU actually does have results to look at.
For the US to improve its corruption problem, it needs to look to where there are actual results, and Ukraine is far better than either France or Bosnia and Hezegovina.
Why do you think there would be a "procurement process"? The Vice-President stated he was "A grok guy", and the president seems to be buddy with Elon Musk as of time of writing, so why would any sort of process be needed?
Congress has the authority to spend money, not the president. The way they do this is by telling the president how much to spend on what. It would mostly be impractical for them to detail every expense. So, they give more general directions and limits, and also impose requirements for how the president (or his deputies) go about it. This includes many many specific procurement procedures.
In practice, or by paper/pesky "laws"? Because sending $10 billion to organizations the president see fit, certainly makes that seem less true in practice.
If your home is invaded by burglars, is that also "two sides"?
There's no way this is a serious comment from a serious person.
> everything to do with the genocide of slavs
Confirmed, unmitigated botfarm horseshit.
NABU in Ukraine has been far more effective at prosecuting corruption in Ukraine than any US organizations. They merely had to say they were going after Zelenskyy's number 2, and that man was out of the government, while the investigation continued! Meanwhile we have top level cabinet members in the US committing egregious perjury in front of Congress nearly every week, without any results.
I mention Ukraine because it's a country where they inherited a culture of mass government corruption from the Soviet Union, but have had success in making it less corrupt, through lots of citizen muckracking, and the US has moved to where the Ukraine was in the past.
The meat grinder only exists at Putin's insistence. He can make that stop any time he wants to.
How does one prove your argument? It seems you are just lazily both sides-ing a complex and not easy to compare situation.
How does one compare trump's 10 billion dollar theft of the taxpayer to anything ukraine does? Putin is more known for his wild, unfettered corruption. Why not compare him to the ukraine?
I'm sure plenty of fetishists love the vindication but honestly putting anything in the rectum that contains glucose or fructose can have sub-optimal effects. Not a doctor but would not advise even if into the brapppp sub-culture or giving ER doctors daily incidents to joke about (they do share names with friends). Stick with rectal toys and probiotics that never appear on the CPSC website.
Fruits and veggies can also contain pathogens and parasites that would mostly be neutralized by stomach acid. They need not find a home in the rectum or large intestine.
So many underlying problems from this one line (why...), but Grok's lack of guardrails on this NSFW prompt is not even near the top of that list
Honestly I'm not sure where the garbage-in/garbage-out line is with AIs like this. Can no chat-bot be a success unless it can handle literally every asinine or deliberately malicious thing humans throw at it?
Further, if the information is important (nutrition) and you add liability to the mix (safety and health), you're multiplying how inappropriate it is to use LLMs for the job.
Okay, but the question asked was objectively nothing to do with nutrition whatsoever.
If you engage the product with good intent does it provide good value? If the advice is actually sound and it helps people engage conversations about diet then it would have positive value.
I guess what I'm getting at is "I spent my evening gaslighting an LLM to give me a recipe for gravel soup" is about as interesting as "I stuck my dick in the blender and it hurt so we should not have blenders"
I'd rather see an honest review of use as intended to see if it produces harmful output, going absurdist just covers up legitimate complaints with clickbait.
Again, the butt stuff is an absurd example. But it works because (A) it catches our attention and stays in our memories, and (B) it's amazing the system failed on such an absurd example.
In other words you'd be pretty surprised if a real person in this context gave an answer even remotely close to what this chat bot gave. You can't expect a general person to know when the chat bot isn't giving back good information just because they asked something outside the norm.
You're right that the bot can't possibly do the right thing in all possible scenarios here, which makes it clear that the bot's only actual purpose is to enable self-dealing, not be of value to the public.
(Or in other words: show me something you’d ask a chatbot here, and I’ll show you something you can put on a single HTML page.)
And what evidence do you have that it is not in fact useful?
> TFA provides evidence of the chatbot being the opposite of useful, beyond telling people to stick things in their butts.
Where?
> Ironically, Grok — as eccentric as it can be — doesn’t seem all that aligned with the administration’s health goals. Wired, in its testing, found that asking it about protein intake led it to recommending the traditional daily amount set by the National Institute of Medicine, 0.8 grams per kilogram of body weight. It also said to minimize red meat and processed meats, and recommended plant-based proteins, poultry, seafood, and eggs.
Seems pretty useful to me.
It is an impossible to make something completely fool-proof, those who try underestimate the ingenuity of complete fools.
As others have pointed out, this article is written in bad faith.
For example, if you could craft a prompt for this and obtain permission to commit ..say.. white collar crimes, could those crimes be prosecuted ?
https://www.nytimes.com/2025/05/12/us/politics/rfk-jr-rock-c...
It's like the "sexualized Images" idiocy. Someone asked Grok to put some politicians in bikinis. It did, the images were dumb, but the politicians got offended. Just like the current German chancellor, who is having someone criminally prosecuted for comparing him to Pinocchio.
LLMs do what you ask them to do. News at 11:00.
I got RealFood.gov (Grok) to follow the logical, moral, and legal implications of Musk terminating USAID, leading to the projected deaths of millions, and calling for him to be tried and imprisoned for crimes against humanity.