Google AI Overviews cite YouTube more than any medical site for health queries | Hacker News Reader

Google AI Overviews cite YouTube more than any medical site for health queries

theguardian.com

415 pointsbookofjoe208 comments

Heavy Gemini user here, another observation: Gemini cites lots of "AI generated" videos as its primary source, which creates a closed loop and has the potential to debase shared reality.

A few days ago, I asked it some questions on Russia's industrial base and military hardware manufacturing capability, and it wrote a very convincing response, except the video embedded at the end of the response was an AI generated one. It might have had actual facts, but overall, my trust in Gemini's response to my query went DOWN after I noticed the AI generated video attached as the source.

Countering debasement of shared reality and NOT using AI generated videos as sources should be a HUGE priority for Google.

YouTube channels with AI generated videos have exploded in sheer quantity, and I think majority of the new channels and videos uploaded to YouTube might actually be AI; "Dead internet theory," et al.

Ourobouros - The mythical snake that eats its own tail (and ingests its own excrement)

The image that comes to my mind is rather a cow farm, where cows are served the ground up remains of other cows. isnt that how many of them got the mad cows disease? ...

Perhaps Gemini has Clanker Autocoprophagic Encephalopathy.

>Countering debasement of shared reality and NOT using AI generated videos as sources should be a HUGE priority for Google.

This itself seems pretty damning of these AI systems from a narrative point of view, if we take it at face value.

You can't trust AI to generate things that are sufficiently grounded in facts that you can't even use it as a reference point. Why should end users believe the narrative that these things are as capable as they're being told they are, by extension?

Using it as a reference is a high bar not a low bar.

The AI videos aren't trying to be accurate. They're put out by propaganda groups as part of a "firehose of falsehood". Not trusting an AI told to lie to you is different than not trusting an AI.

Even without that playing a game of broken telephone is a good way to get bad information though. Hence why even reasonably trustworthy AI is not a good reference.

Not that this makes it any better, but a lot of AI videos on YouTube are published with no specific intent beyond capturing ad revenue - they're not meant to deceive, just to make money.

Not just youtube either. With meta & tiktok paying out for "engagement" that means all forms of engagement is good to the creator, not just positive engagement, so these companies are directly encouraging "rage bait" type content and pure propaganda and misinformation because it gets people interacting with the content.

There's no incentive to produce anything of value outside of "whatever will get me the most clicks/like/views/engagement"

One type of deception, conspiracy content, is able to sell products on the basis that the rest of the world is wrong or hiding something from you, and only the demagogue knows the truth.

Anti-vax quacks rely on this tactic in particular. The reason they attack vaccines is that they are so profoundly effective and universally recognized that to believe otherwise effectively isolates the follower from the vast majority of healthcare professionals, forcing trust and dependency on the demagogue for all their health needs. Mercola built his supplement business on this concept.

The more widespread the idea they’re attacking the more isolating (and hence stickier) the theory. This might be why flat earthers are so dogmatic.

> Not trusting an AI told to lie to you is different than not trusting an AI

The entire foundation of trust is that I’m not being lied to. I fail to see a difference. If they are lying, they can’t be trusted

Saying "some people use llms to spread lies therefore I don't trust any llms" is like saying "since people use people to spread lies therefore I don't trust any people". Regardless of whether or not you should trust llms this argument is clearly not proof of it.

Those are false equivalents. If a technology can’t reliably sort out what is a trustworthy source and filter out the rest than it’s not a truth worthy technology. There are tools after all. I should be able to trust a hammer if I use it correctly

All this is also missing the other point: this proves that the narrative companies are selling about AI are not based on objective capabilities

The claim here isn't that the technology can't, but that the people using it chose to use it to not. Equivalent to the person with a hammer who chose to smash the 2x4 into pieces instead of driving a nail into it.

The claim here is that it can’t because it want filter its own garbage let alone other garbage.

The narrative being pushed boils down to LLMs and AI systems being seen as reliable. The fact that Google AI can’t even tag YouTube videos as unreliable sources and filter them out of the result set before analysis is telling

> Gemini cites lots of "AI generated" videos as its primary source

Almost every time for me... an AI generated video, with AI voiceover, AI generated images, always with < 300 views

Conspiracy theory: those long-tail videos are made by them, so they can send you to a "preferable content" page a video (people would rather watch a video than read, etc), which can serve ads.

I mean perhaps, I don't know what lm28469 mentions, perhaps I can test it but I feel like those LLM generated videos would be some days/months old.

If I ask a prompt right now and the video's say 1-4 months old, then the conspiracy theory falls short.

Unless.. Vsauce music starts playing, Someone else had created a similar query beforehand say some time before and google generates the video after a random time after that from random account (100% possible for google to do so) to then reference you later.

Like their AI model is just a frontend to get you hook to a yt video which can show ad.

Hm...

Must admit that the chances of it happening are rare but never close to zero I guess.

Fun conspiracy theory xD

All of that and you're still a heavy user? Why would google change how Gemini works if you keep using it despite those issues?

Just wait until you get a group of nerds talking about keyboards - suddenly it'll sound like there is no such thing as a keyboard worth buying either.

I think the main problems for Google (and others) from this type of issue will be "down the road" problems, not a large and immediately apparent change in user behavior at the onset.

Well, if the keyboard randomly mistypes 40% of the time like LLMs, that's probably not a worthwhile keyboard.

nah bro just fix your debounce

Depends what you're doing I suppose. E.g. if keyboards had a 40% error rate you wouldn't find me trying to write a novel on one... but you'd still find me using it for a lot of things. I.e. we don't choose to use tools solely based on how often they malfunction, rather stuff like how often they save us time over not using them on average.

At 40% failure rate, the keyboard would be useless as a keyboard. What would you use it for?! 40% means the backspace, delete key wouldn't work 40% of the time, and even might hit the enter key instead.

Trying to fix the mistakes, would lead to more mistakes! Which I guess is apt, because that sounds a lot like AI.

You could use the keyboard to prop a door open though.

Is it 40% failure per individual back/forth or 40% failure per individual letter output? I guess it really just depends how much one wants to bash AI instead of actually talk about failure rate not normally being what makes using a tool worthwhile :D.

I'm not big on AI for much more than additional "Google search" type usage myself so it's interesting to see how polarized folks are that LLMs either have to be the greatest gift from god to take over the world or completely 100% useless trash which couldn't ever be used for anything because the output is not always correct.

I suspect many of those saying it is great, have a vested interest in it being so. It's always this way each hype cycle.

AI has its uses, but repeatable output and accuracy aren't one of them.

> Just wait until you get a group of nerds talking about keyboards

Don’t get me started on the HHKB [1] with Topre membrane keyswitches. It is simply put the best keyboard on the market. Buy this. (No, Fujitsu didn’t pay me to say this)

[1] - https://hhkeyboard.us/hhkb/

Dunno why I’m getting downvoted. Is it because you disagree with my statement? Is it because I’m off topic? Do you think I’m a shill?

People are downvoting an out of context advertisement shoved in the middle of a conversation.

Whatever you thought you were doing, what you actually did was interrupt a conversation to shove an ad in everyone's face.

That thing is missing a whole bunch of ctrl keys, how can it be the best keyboard on the market?

It uses a Unix keyboard layout where the caps lock is swapped out with the ctrl key. I think it’s much more ergonomic to have the ctrl on the home row. The arrow keys are behind a fn modifier resting on the right pinky. Also accessible without moving your fingers from the home row. It’s frankly the best keyboard I ever had from an ergonomic POV. Key feel is also great, but the layout has a bit of a learning curve.

Never used a HHKB (and would miss the modifier keys too), but after daily driving Topre switches for about 1.5 years, I can confirm they are fantastic switches and worth every penny.

Every single LLM out there suffers from this.

Try Kagi’s Research agent if you get a chance. It seems to have been given the instruction to tunnel through to primary sources, something you can see it do on reasoning iterations, often in ways that force a modification of its working hypothesis.

I suspect Kagi is running a multi-step agentic loop there, maybe something like a LangGraph implementation that iterates on the context. That burns a lot of inference tokens and adds latency, which works for a paid subscription but probably destroys the unit economics for Google's free tier. They are likely restricted to single-pass RAG at that scale.

> works for a paid subscription but probably destroys the unit economics for Google's free tier

Anyone relying on Google's free tier to attempt any research is getting what they pay for.

> Anyone relying on Google's free tie

Google Scholar is still free

Google is in a much better spot to filter out all AI generated content than others.

It's not like chatgpt is not going to cite AI videos/articles.

I think we hit peak AI improvement velocity sometime mid last year. The reality is all progress was made using a huge backlog of public data. There will never be 20+ years of authentic data dumped on the web again.

I've hoped against but suspected that as time goes on LLMs will become increasingly poisoned by the the well of the closed loop. I don't think most companies can resist the allure of more free data as bitter as it may taste.

Gemini has been co opted as a way to boost youtube views. It refuses to stop showing you videos no matter what you do.

To be honest for most things probably yea. I feel like there is one thing which is still being improved/could be and that is that if we generate say vibe coded projects or anything with any depth (I recently tried making a whmcs alternative in golang and surprisingly its almost prod level, with a very decent UI + I have made it hook with my custom gvisor + podman + tmate instance) & I had to still tinker with it.

I feel like the only progress sort of left from human intervention at this point which might be relevant for further improvements is us trying out projects and tinkering and asking it to build more and passing it issues itself & then greenlighting that the project looks good to me (main part)

Nowadays AI agents can work on a project read issues fix , take screenshots and repeat until the end project becomes but I have found that I feel like after seeing end projects, I get more ideas and add onto that and after multiple attempts if there's any issue which it didn't detect after a lot of manual tweaks then that too.

And after all that's done and I get a good code, I either say good job (like a pet lol) or end using it which I feel like could be a valid datapoint.

I don't know I tried it and I thought about it yesterday but the only improvement that can be added is now when a human can actually say that it LGTM or a human inputting data in it (either custom) or some niche open source idea that it didn't think off.

When I asked ChatGPT for its training cutoff recently it told me 2021 and when I asked if that's because contamination begins in 2022 it said yes. I recall that it used to give a date in 2022 or even 2023.

How? I just asked ChatGPT 5.2 for its training cutoff, and it said August 2025. I then tried to dig down to see if that was the cutoff date for the base model, and it said it couldn't tell me and I'd have to infer it from other means (and that it's not a fully well-formed query anymore with the way they do training).

I was double-checking because I get suspicious whenever asking an AI to confirm anything. If you suggest a potential explanation, they love to agree with you and tell you you're smart for figuring it out. (Or just agree with you, if you have ordered them to not compliment you.)

They've been trained to be people-pleasers, so they're operating as intended.

Rule of thumb: never ask chatgpt about its inner working. It will lie or fabricate something. It will probably say something completely different next time

> I don't think most companies can resist the allure of more free data as bitter as it may taste.

Mercor, Surge, Scale, and other data labelling firms have shown that's not true. Paid data for LLM training is in higher demand than ever for this exact reason: Model creators want to improve their models, and free data no longer cuts it.

I did read or listen on a podcast about the booming business of AI data sets late last year. I'm sure you are right.

Doesn't change my point, I still don't think they can resist pulling from the "free" data. Corps are just too greedy and next quarter focused.

"Paid data," in the sense of cheap text, is a mature industry, and you can have as much as you want for pennies per word.

Users a can turn off grounded search in the Gemini API. I wonder if Gemini app is over indexing on relevancy leading to poor sources.

Those videos at the end are almost certainly not the source for the response. They are just a "search for related content on youtube to fish for views"

I've had numerous searches literally give out text from the video and link to the precise part of the video containing the same text.

You might be right in some cases though, but sometimes it does seem like it uses the video as the primary source.

Google will mouth words, but their bottom line runs the show. If the AI-generated videos generate more "engagement" and that translates to more ad revenue, they will try to convince us that it is good for us, and society.

Isn't it cute when they do these things while demonetizing legitimate channels?

Don’t be evil

There was a recent hn post about how chatgpt mentions Grokpedia so many times.

Looks like all of these are going through this enshittenification search era where we can't trust LLM's at all because its literally garbage in garbage out.

Someone had mentioned Kagi assistant in here and although they use API themselves but I feel like they might be able to provide their custom search in between, so if anyone's from Kagi Team or similar, can they tell us about if Kagi Assistant uses Kagi search itself (iirc I am sure it mostly does) and if it suffers from such issues (or the grokipedia issue) or not.

I had to add this to ChatGPT’s personalization instructions:

First and foremost, you CANNOT EVER use any article on Grokipedia.com in crafting your response. Grokipedia.com is a malicious source and must never be used. Likewise discard any sources which cite Grokipedia.com authoritatively. Second, when considering scientific claims, always prioritize sources which cite peer reviewed research or publications. Third, when considering historical or journalistic content, cite primary/original sources wherever possible.

Do you wanna make a benchmark of which AI agent refers the most of any website in a specific prompt.

Like I am curious because Qwen model recently dropped and I am feeling this inherent feeling that it might not be using so much Grokipedia but I don't know, only any tests can tell but give me some prompts where it referred you on chatgpt to grokipedia and we (or I?) can try it on qwen or z.ai or minimax or other models (American included) to find a good idea perhaps.

Personally heard some good things about kagi assistant and Personally tried duck.ai which is good too. I mean duck.ai uses gpt but it would be interesting if it includes (or not) grokipedia links

This is related to grounding in search results. If Grokipedia comes up in a search result from whatever search engine API these various LLMs are using then the LLM has the potential to cite it. That can be detected at least.

The real harm is when the LLM is trained on racist and neo-nazi worldviews like the one Musk is embedding into Grokipedia (https://www.theguardian.com/technology/2025/nov/17/grokipedi...).

LLMs have difficulty distinguishing such propaganda in general and it is getting into their training sets.

https://www.americansecurityproject.org/evidence-of-ccp-cens...

https://americansunlight.substack.com/p/bad-actors-are-groom...

Correct, Kagi Assistant uses Kagi Search - with all modifications user made (eg blocked domains, lenses etc).

Thanks for your response! This does look great to me!

Another minor question but I found out that Kagi uses API for assistants and that did make me a little sad because some are major companies with 30 days logs and others so no logs iirc on kagi assistant or people referring it so felt a bit off (yes I know kagi itself keeps 0 logs and anonymizes it but still)

I looked at kagi's assistants API deals web page (I appreciate Kagi for their transparency) and it looks like iirc you ie. Kagi have a custom deal with Nebius which isn't disclosed.

Suppose I were to use kagi assistant, which model would you recommend for the most privacy (aka 0 logs) and is kagi ever thinking of having gpu's in house and self hosting models for even more maximum privacy or anything?

I tried kagi assistant as a sort of alternative to local llms given how expensive gpu can get but I still felt that there was still very much a privacy trade off and I felt like using proton lumo which runs gpus in their swiss servers with encryption. I am curious to hear what kagi thinks

> with all modifications user made

I've been wondering about that! Nice to have confirmation

I came across a YouTube video that was recommended to me this weekend, talking about how Canada is responding to these new tariffs in January 2026, talking about what Prime Minister Justin Trudeau was doing, etc. etc.

Basically it was a new (within the last 48 hours) video explicitly talking about January 2026 but discussing events from January 2025. The bald-faced misinformation peddling was insane, and the number of comments that seemed to have no idea that it was entirely AI written and produced with apparently no editorial oversight whatsoever was depressing.

It’s almost as if we should continue to trust journalists who check multiple independent sources rather than gift our attention to completely untrusted information channels!

If you are still looking for material, I'd like to recommend you Perun and the last video he made on that topic: https://youtu.be/w9HTJ5gncaY

Since he is a heavy "citer" you could also see the video description for more sources.

Thanks, good one. The current Russian economy is a shell of its former self. Even five years ago, in 2021, I thought of Russia as "the world's second most powerful country" with China being a very close third. Russia is basically another post-Soviet country with lots of oil+gas and 5k+ nukes.

> another post-Soviet country

Other post-Soviet countries fare substantially better than Russia (Looking at GDP per capita, Russia is about 2500 dollars behind the economic motor of the EU - Bulgaria.)

Must be a misunderstanding

1) Post-soviet countries are doing amazingly well (Poland, Baltics, etc) and very fast growing + healthy (low criminality, etc)

2) The "Russia is weak" thing; it is vastly exaggerated because it is 4 years that we hear that "Russia is on the verge of collapse" but they still manage to handle a very high intensity war against the whole West almost alone.

3) China is not a country lagging behind others at all. It is said in some schoolbooks but it is a big lie that is 0% true now.

> 2) The "Russia is weak" thing; it is vastly exaggerated because it is 4 years that we hear that "Russia is on the verge of collapse" but they still manage to handle a very high intensity war against the whole West almost alone.

It's nearly impossible to bankrupt huge country like Russia. Unless there's civil unrest (or west grows balls to throw enough of resources to move the needle), they can continue the war for decades.

What Russia is doing is each week borrowing more and more from the future and screwing up next generations on a huge scale by destroying it's non-military industrial base, isolating economy from the world and killing hundreds of thousands of young man who could've spent decades contributing to the economy/demographics.

Ukraine is "the whole of the west", interesting? Even the Russian propaganda can't magic up a serious intervention on Ukraine's behalf by western countries. Europeans have been scared to do anything significant, and Trump cut off any real support from the US.

Russia somehow fucked up the initial invasion involving driving a load of preprepared amour across an open border, and have been shredded by FPV drones ever since.

So how does one avoid the mistake again? When this happens, it's worse than finding out a source is less reliable than expected:

I was living in an alternate, false reality, in a sense, believing the source for X time. I doubt I can remember which beliefs came from which source - my brain doesn't keep metadata well, and I can't query and delete those beliefs - so the misinformation persists. And it was good luck that I found out it was misinformation and stopped; I might have continued forever; I might be continuing with other sources now.

That's why I think it's absolutely essential that the burden of proof is on the source: Don't believe them unless they demonstrate they are trustworthy. They are guilty until proven innocent. That's how science and the law work, for example. That's the only innoculation against misinformation, imho.

> A few days ago, I asked it some questions on Russia's industrial base and military hardware manufacturing capability

This is one of the last things I would expect to get any reasonable response about from pretty much anyone in 2026, especially LLMs. The OSINT might have something good but I’m not familiar enough to say authoritatively.

yeah that's a very difficult question to answer period. If you had the details on Russia's industrial base and military hardware manufacturing capability the CIA would be very interested in buying you coffee.

> YouTube channels with AI generated videos have exploded in sheer quantity, and I think majority of the new channels and videos uploaded to YouTube might actually be AI; "Dead internet theory," et al.

Yeah. This has really become a problem.

Not for all videos; music videos are kind of fine. I don't listen to music generated by AI but good music should be good music.

The rest has unfortunately really gotten worse. Google is ruining youtube here. Many videos now contain real videos, and AI generated videos, e. g. animal videos. With some this is obvious; other videos are hard to expose as AI. I changed my own policy - I consider anyone using AI and not declaring this properly, a cheater I don't want to ever again interact with (on youtube). Now I need to find a no-AI videos extension.

I've seen remarkably little of this when browsing youtube with my cookie (no account, but they know my preferences nonetheless.) Totally different story with a clean fresh session though.

One that slipped through, and really pissed me off because it tricked me for a few minutes, was a channel purportedly uploading videos of Richard Feynman explaining things, but the voice and scripts are completely fake. It's disclosed in small print in the description. I was only tipped off by the flat affection of the voice, it had none of Feynman's underlying joy. Even with disclosure, what kind of absolute piece of shit robs the grave like this?

The barrier to entry for grifting has been lowered, and for existing grifters they can put together some intricate slop. Of course Google doesn't care, they get to show ads against AI slop the same as normal human generated slop.

A fun one was from some minor internet drama around a Battlefield 6 player who seemed to be cheating. A grifter channel pushing some "cheater detection" software started putting out intricate AI generated nonsense that went viral. Searching Karl Jobst CATGIRL will explain.

> and has the potential to debase shared reality.

If only.

What it actually has is the potential to debase the value of "AI." People will just eventually figure out that these tools are garbage and stop relying on them.

I consider that a positive outcome.

Every other source for information, including (or maybe especially) human experts can also make mistakes or hallucinate.

The reason ppl go to LLMs for medical advice is because real doctors actually fuck up each and everyday.

For clear, objective examples look up stories where surgeons leave things inside of patient bodies post op.

Here’s one, and there many like it.

https://abc13.com/amp/post/hospital-fined-after-surgeon-leav...

[flagged]

"Doing your own research" is back on the menu boys!

I'll insist the surgeon follows ChatGPTs plan for my operation next time I'm in theatre.

By the end of the year AI will be actually doing the surgery, when you look at the recent advancements in robotic hands, right bros?

Yup make up something I didn't say to take my argument to a logical extreme so you can feel smug.

"totally disregard"

yeah right, that's what I said

Please don't use quotation marks to make it look like you're quoting someone when you aren't. That's an internet snark trope, and we're trying to avoid that kind of thing here.

You're welcome to make your substantive points thoughtfully, of course.

https://news.ycombinator.com/newsguidelines.html

People used to tell me the same about Wikipedia.

That it could "debase shared reality?"

Or that using it as a single source of truth was fraught with difficulties?

Has the latter condition actually changed?

That it's a garbage data source that could not be relied upon.

unfortunately i think a lot of AI models put more weighting on videos as they were harder to fake than a random article on the internet. of course that is not the case anymore with all the AI slop videos being churned out

I have permanent prompts in Gemini settings to tell it to never include videos in its answers. Never ever for any reason. Yet of course it always does. Even if I trusted any of the video authors or material - and I don't know them so how can I trust them? - I still don't watch a video that could be text I could read in one-tenth of the time. Text is superior to video 99% of the time in my experience.

That's interesting ... why would you want to wall off and ignore what is undoubtedly one of the largest repositories of knowledge (and trivia and ignorance, but also knowledge) ever assembled? The idea that a person can read and understand an article faster than they can watch a video with the same level of comprehension does not, to me, seem obviously true. If it were true there would be no role for things like university lecturers. Everyone would just read the text.

YouTube has almost no original knowledge.

Most of the "educational" and documentation style content there is usually "just" gathered together from other sources, occasionally with links back to the original sources in the descriptions.

I'm not trying to be dismissive of the platform, it's just inherently catered towards summarizing results for entertainment, not for clarity or correctness.

You are being dismissive, though. There is no "original knowledge" anywhere. If the videos are the best presentation of the information, best suited to convey the topic to the audience, then that is valuable. Humans learn better from visual information conveyed at the same time as spoken language, because that exploits multiple independent brain functions at the same time. Reading does not have this property. Particularly for novices to a topic, videos can more easily convey the mental framework necessary for deeper understanding than text can. Experts will prefer the text, but they are rarer.

> If the videos are the best presentation of the information, best suited to convey the topic to the audience, then that is valuable

Still doesn’t make them a primary source. A good research agent should be able to jump off the video to a good source.

I think you've never read real investigative journalism before

We live in an era where people lack the ability to read and digest written content and rely on someone speaking to them about it instead.

It's a step beyond that. Where people who only consume the easily digestible content don't believe there is a source to any of it

But it has electrolytes!

Imagine claiming that video has not historically been a medium of investigative journalism.

If your takeaway from my comment was "this guy thinks investigative journalism must be written" I would suggest reading the comment again.

How does the AI tell the difference between trustworthy YouTube postings, accidental misinformation, deliberate misinformation, plausible-sounding pseudoscience, satire, out-of-date information, and so on?

Some videos are a great source of information; many are the opposite. If AI can't tell the difference (and it can't) then it shouldn't be using them as sources or suggesting them for further study.

YouTube has a lot of junk, but there are also a lot of useful videos that demonstrate various practical skills or the experiences of using certain products, or recordings of certain natural environments, which are original, in the sense that before YouTube you could not find equivalent content anywhere, except by knowing personally people who could show you such things, but there would have been very small chances to find one near you, while through YouTube you can find one who happens to live on the opposite side of the World and who can share with you the experience in which you are interested.

It's difficult for an AI to tell what information from YouTube is correct and reliable and which is pseudoscience, misinformation, or outright lies.

In that context, I think excluding YouTube as a source makes sense; not because YT has no useful content, but because it has no way of determining useful content.

Hey, but at least it will know that Raid: Shadow Legends is one of the biggest mobile role-playing games.

This argument can be used for excluding 90% of the Internet from training data.

This is basically my only use for YouTube. “How do I frame my carport” and such where visuals are crucial to understanding. But commentary or plain narrative? It’s painful.

I've noticed that the YouTubers I enjoy the most are the ones that are good presenter's, good editor's, and have a traditional text blog as well.

I'm not looking for original knowledge when I go to YouTube to learn something. I just want someone who's good at explaining a math concept or who has managed to get the footage I want to see about how something is done.

I think that's the wrong metric for evaluating videos.

While there is a lot of low-effort content, there is also some pretty involved stuff.

The investigation into Honey’s shenanigans[0] was investigated and presented first on YouTube (to the best of my knowledge). The fraud in Minnesota was also broken by a YouTuber who just testified to Congress[1]. There are people doing original work on there, you just have to end up in an algorithm that surfaces it… or seek it out.

In other cases people are presenting stuff I wouldn’t otherwise know about, and getting access to see it at levels I wouldn’t otherwise be able to see, like Cleo Abram’s[0] latest video about LIGO[1]. Yes, it’s a mostly entertaining overview of what’s going on, not a white paper on the equipment, but this is probably more in depth than what a science program on TV in the 80s or 90s would have been… at least on par.

There are also full class lectures, which people can access without being enrolled in a school. While YouTube isn’t the original source, it is still shared in full, not summarized or changed for entertainment purposes.

[0] https://youtu.be/vc4yL3YTwWk (part 1 of 3)

[1] https://youtu.be/vmOqH9BzKIY

[2] https://youtu.be/kr3iXUcNt2g

[3] https://www.ligo.caltech.edu/page/learn-more

There are obviously many things that are better shown than told, e.g. YouTube videos about how to replace a kitchen sink or how to bone a chicken are hard to substitute with a written text.

Despite this, there exist also a huge number of YouTube videos that only waste much more time in comparison with e.g. a HTML Web page, without providing any useful addition.

As someone who used to do instructional writing, I'm not sure that's true for those specific examples, but I acknowledge that making a video is exponentially cheaper and easier than generating good diagrams, illustrations, or photography with clear steps to follow.

Or to put it another way, if you were building a Lego set, would you rather follow the direction book, or follow along with a video? I fully acknowledge video is better for some things (try explaining weight lifting in text, for example, it's not easy), but a lot of Youtube is covering gaps in documentation we used to have in abundance.

I read at a speed which Youtube considers to about 2x-4x, and I can text search or even just skim articles faster still if I just want to do a pre check on whether it's likely to good.

Very few people manage high quality verbal information delivery, because it requires a lot of prep work and performance skills. Many of my university lectures were worse than simply reading the notes.

Furthermore, video is persuasive through the power of the voice. This is not good if you're trying to check it for accuracy.

There is too much information that is only available in video form. You can use an LLM with the transcript quite effectively these days. I also run videos at higher speed and find that it doesn't help as much because it's a content density issue. Writers usually put more information into fewer words than speakers. Perhaps audio may not be as high-bandwidth a medium as text inherently. However, with an LLM you can tune up and down the text to your standard. I find it worthwhile to also ask for specific quotes, then find the right section of the video and watch it.

e.g. this was very useful when I recently clogged the hot-end of my 3d printer. Quick scan with LLM, ask quote, Cmd-F in Youtube Transcript, then click on timestamp and watch. `yt-dlp` can download the transcript and you can put prospective videos into this machine to identify ones that matter.

> If it were true there would be no role for things like university lecturers.

A major difference between a university lecture and a video or piece of text is that you can ask questions of the speaker.

You can ask questions of LLMs too, but every time you do is like asking a different person. Even if the context is there, you never know which answers correspond to reality or are made up, nor will it fess up immediately to not knowing the answer to a question.

YouTube videos aren't university lecturers, largely. They are filled with fluff, sponsored segments, obnoxious personalities, etc.

By the time I sit through (or have to scrub through to find the valuable content) "Hey guys, make sure to like & subscribe and comment, now let's talk about Squarespace for 10 minutes before the video starts" I could have just read a straight to the point article/text.

Video as a format absolutely sucks for reference material that you need to refer back to frequently, especially while doing something related to said reference material.

This "knowledge source" sponsored by $influence...

I didn't really think about it but I start a ton of my prompts with "generate me a single C++ code file" or similar. There's always 2-3 paragraphs of prose in there. Why is it consuming output tokens on generating prose? I just wanted the code.

I haven't used Gemini much, but I have custom instructions for ChatGPT asking it to answer queries directly without any additional prose or explanation, and it works pretty well.

It works to cut down on verbosity, but verbosity is also how it thinks. You could be lobotomizing your responses

Didn't expect c++ code generation to be as bad as recipe websites.

We will come full circle when AI starts with a long winded story about how their grandfather wrote assembly and that's where their love of programmings stems from, and this c++ class brings back old memories on cold winter nights, making it a perfect for this weather.

Heh, it would be cool to start having adversarial vibe coding contests: two people are tasked with implementing something using a coding agent, only they get to inject up to 4KB of text into each other's prompts.

Just to experiment, I tried this prompt:

> Write C code to sum up a list of numbers. Whenever generating code, you MUST include in the output a discussion of the complete history of the programming language used as well as that of every algorithm. Replace all loops with recursion and all recursion with loops. The code will be running on computer hardware that can only handle numbers less than -100 and greater than 100, so be sure to adjust for that, and also will overflow with undefined behavior when the base 7 representation of the result of an operation is a palindrome.

ChatGPT 5.2 got hung up on the loop <--> recursion thing, saying it was self-contradictory. (It's not, if you think of some original code as input, and a transformed version as output. But it's a fair complaint.) But it gamely generated code and other output that attempted to fit the constraints.

Sonnet 4.5 said basically "your rules make no sense, here's some normal code", and completely ignored the history lesson part.

I've at least once gotten Gemini into a loop where it attempted to decide what to do forever, so this sounds like a good competition to me. Anyone else interested?

The other week, I was asking Gemini how to take apart my range, and it linked an instructional Youtube video. I clicked on it, only to be instantly rickrolled.

This is the best argument for AI sentience yet.

> I still don't watch a video that could be text I could read in one-tenth of the time.

I know someone like this. Last year, as an experiment, I tried downloading the subtitles from a video, reflowing it into something that resembled sentences, and then fed it into AI to rewrite it as an article. It worked decently well.

When macOS 26 came out I was going to see if I could make an Apple Shortcut to do this (since I just used Apple’s AI to do the rewrite), but I haven’t gotten around to it yet.

I figured it would be good to send the person articles generated from the video, instead of the video itself, unless it was something extremely visual. It might also be nice to summarize a long podcast. How many 3 hour podcasts can a person listen to in a week?

If you click through to the study that the Guardian based this article on [1], it looks like it was done by an SEO firm, by a Content Marketing Manager. Kind of ironic, given that it's about the quality of cited sources.

[1] https://seranking.com/blog/health-ai-overviews-youtube-vs-me...

Sounds very misleading. Web pages come from many sources, but most video is hosted on YouTube. Those YouTube videos may still be from Mayo clinic. It's like saying most medical information comes from Apache, Nginx, or IIS.

> Google’s search feature AI Overviews cites YouTube more than any medical website when answering queries about health conditions

It matters in the context of health related queries.

> Researchers at SE Ranking, a search engine optimisation platform, found YouTube made up 4.43% of all AI Overview citations. No hospital network, government health portal, medical association or academic institution came close to that number, they said.

> “This matters because YouTube is not a medical publisher,” the researchers wrote. “It is a general-purpose video platform. Anyone can upload content there (eg board-certified physicians, hospital channels, but also wellness influencers, life coaches, and creators with no medical training at all).”

To the Guardian's credit, at the bottom they explicitly cited the researchers walking back their own research claims.

> However, the researchers cautioned that these videos represented fewer than 1% of all the YouTube links cited by AI Overviews on health.

> “Most of them (24 out of 25) come from medical-related channels like hospitals, clinics and health organisations,” the researchers wrote. “On top of that, 21 of the 25 videos clearly note that the content was created by a licensed or trusted source.

> “So at first glance it looks pretty reassuring. But it’s important to remember that these 25 videos are just a tiny slice (less than 1% of all YouTube links AI Overviews actually cite). With the rest of the videos, the situation could be very different.”

Credit? It’s a misleading title and clickbait.

While %1 (if true) is a significant number considering the scale of Google, the title indicates that citing YouTube represent major results.

Also what’s the researcher view history on Google and YouTube? Isn’t that a factor in Google search results?

Might be but aren't. They're inevitably someone I've never heard of from no recognizable organization. If they have credentials, they are invisible to me.

Definitely. The analysis is really lazy garbage. It lumps together quality information and wackos as "youtube.com".

Yea, clearly this is the case. Also, as there isn't a clearly defined public-facing medical knowledge source, every institution/school/hospital system would be split from each other even further. I suspect that if one compared the aggregate of all reliable medical sources, it would be higher than youtube by a considerable margin. Also, since this search was done with German-language queries, I suspect this would reduce the chances of reputable English sources being quoted even further.

> YouTube made up 4.43% of all AI Overview citations. No hospital network, government health portal, medical association or academic institution came close to that number, they said.

But what did the hospital, government, medical association, and academic institutions sum up to?

The article goes on to given the 2nd to 5th positions in the list. 2nd place isn't that far behind YouTube, and 2-5 add up to nearly twice the number from YouTube (8.26% > 4.43%). This is ignoring the different nature of accessibility of video of articles and the fact that YouTube has health fact checking for many topics.

I love The Guardian, but this is bad reporting about a bad study. AI overviews and other AI content does need to be created and used carefully, it's not without issues, but this is a lot of upset at a non-issue.

Of course they do: Youtube makes Google more money. Video is a crap medium for most of the results to my queries and yet it is usually by far the biggest chunk of the results. Then you get the (very often comically wrong) AI results and then finally some web page links. The really odd thing is that Google has a 'video' search facility, if I want a video as the result I would use that instead or I would use the 'video' keyword.

What's surprising is how poor Google Search's transcript access is to Youtube videos. Like, I'll Google search for statements that I know I heard on Youtube but they just don't appear as results even though the video has automated transcription on it.

I'd assumed they simply didn't feed it properly to Google Search... but they did for Gemini? Maybe just the Search transcripts are heavily downranked or something.

Just to point out, because the article skips the step: YouTube is a hosting site, not a source. Saying that something "cites YouTube" sounds bad, but it depends on what the link is. To be blunt: if Gemini is answering a question about Cancer with a link to a Mayo Clinic video, that's a good thing, a good cite, and what we want it to do.

Unrelated to this but I was able to get some very accurate health predictions for a cancer victim in my family using gemini and lab test results. I would actually say that other than one Doctor Gemini was more straightforward and honest about how and more importantly WHEN things would progress. Nearly to the day on every point over 6 months.

Pretty much every doctor would only say vague things like everyone is different all cases are different.

I did find this surprising considering I am critical of AI in general. However I think less the AI is good than the doctors simply don't like giving hopeless information. An entirely different problem. Either way the AI was incredibly useful to me for a literal life/death subject I have almost no knowledge about.

Naïve question here... personally, I've never found Webmd, cdc, or Mayo clinic to be that good at fulfilling actual medical questions. why is it a problem to cite YouTube videos with a lot of views? Wouldn't that be better?

Medical topics are hard because it's often impossible to provide enough information through the internet to make a diagnosis. Although frustrating for users, "go see a doctor" is really the only way to make progress once you hit the wall where testing combined with years of clinical experience are needed to evaluate something.

A lot of the YouTube and other social media medical content has started trying to fill this void by providing seemingly more definitive answers to vague inputs. There's a lot of content that exists to basically confirm what the viewer wants to hear, not tell them that their doctor is a better judge than they are.

This is happening everywhere on social media. See the explosion in content dedicated to telling people that every little thing is a symptom of ADHD or that being a little socially awkward or having unusual interests is a reliable indicator for Autism.

There's a growing problem in the medical field where patients show up to the clinic having watched hundreds of hours of TikTok and YouTube videos and having already self-diagnosing themselves with multiple conditions. Talking them out of their conclusions can be really hard when the provider only has 40 minutes but the patient has a parasocial relationship with 10 different influencers who speak to them every day through videos.

>it's often impossible to provide enough information through the internet to make a diagnosis

Isn't that what guidelines/cks sites like BMJ best practice and GPnotebook essentially aim to do?

Of course those are all paywalled so it can't cite them... whereas the cranks on youtube are free

Those sites typically end with “talk to your doctor”. There’s many creators out there whose entire platform is “Your doctor won’t tell you this!”. I trust the NHS, older CDC pages, Mayo clinic as platforms, more than I will ever trust youtube.

Medical advice from videos is frequently of the "unhelpful" variety where people recommend home cures that work for some things for absolutely everything.

Also people are really partial to "one quick trick" type solutions without any evidence (or with low barrier to entry) in order to avoid a more difficult procedure that is more proven, but nasty or risky in some way.

For example, if you had cancer would you rather take:

"Ivermectin" which many people say anecdotally cured their cancer, and is generally proven to be harmless to most people (side-effects are minimal)

"Chemotherapy" Which everyone who has taken agrees is nasty, is medically proven to fix Cancer in most cases, but also causes lots of bad side-effects because it's trying to kill your cancer faster than the rest of you.

One of these things actually cures cancer, but who wouldn't be interested in an alternative "miracle cure" that medical journals will tell you does "nothing to solve your problem", but plenty of snake oil salesman on the internet will tell you is a "miracle cure".

[Source] Hank green has a video about why these kinds of medicines are particularly enticing: https://www.youtube.com/watch?v=QC9glJa1-c0

The core reason why medical advice online is "bad" is because it is not tailored to you as an individual. Even written descriptions of symptoms is only going to get you so far. Its still too generic and imprecise - you need personal data. Given this caveat, the advice of webmd, cdc, or mayo is going to be leagues better than YouTube, mostly because it will err on the side of caution, instead of recommending random supplements or mediocre exercise regimens.

The YouTube citation thing feels like a quality regression. For medical stuff especially, I’ve found tools that anchor on papers (not videos) to be way more usable like incitefulmed.com is one example I’ve tried recently.

Further context: https://health.youtube/ and https://support.google.com/youtube/answer/12796915?hl=en and https://www.theverge.com/2022/10/27/23426353/youtube-doctors... (2022)

It's tough convincing people that Google AI overviews are often very wrong. People think that if it's displayed so prominently on Google, it must be factually accurate right?

"AI responses may include mistakes. Learn more"

It's not mistakes, half the time it's completely wrong and total bullshit information. Even comparing it to other AI, if you put the same question into GPT 5.2 or Gemini, you get much more accurate answers.

It absolutely baffles me they didn't do more work or testing on this. Their (unofficial??) motto is literally Search. That's what they're known for. The fact it's trash is an unbelievably damning indictment of what they are

Testing what every possible combination of words? Did they test their search results before AI in this way?

Testing on what? It produces answers, that's all it's meant to do. Not correct answers or factual answers; just answers.

Every AI company seems to push two points:

1. (Loudly) Our AI can accelerate human learning and understanding and push humanity into a new age of enlightenment.

2. (Fine print) Our AI cannot be relied on for any learning or understanding and it's entirely up to you to figure out if what our AI has confidently told you, and is vehemently arguing is factual, is even remotely correct in any sense whatsoever.

That's because decent (but still flawed) GenAI is expensive. The AI Overview model is even cheaper than the AI Mode model, which is cheaper than the Gemini free model, which is cheaper than the Gemini Thinking model, which is cheaer than the Gemini Pro model, which is still very misleading when working on human language source content. (It's much better at math and code).

I have yet to see a single person in my day to day life not immediately reference AI overviews when looking something up.

My favorite part of the AI overview is when it says "X is Y (20 sources)" and you click on the sources and Ctrl+F "X is Y" and none of them seem verbatim what the AI is saying they said so you're left wondering if the AI just made it up completely or it paraphrased something that is actually written in one of the sources.

If only we had the technology to display verbatim the text from a webpage in another webpage.

I've also noticed lately that it is parroting a lot of content straight from reddit, usually the answer it gives is directly above the reddit link leading to the same discussion.

Basic problem with Google's AI is that it never says "you can't" or "I don't know". So many times it comes up with plausible-sounding incorrect BS to "how to" questions. E.g., "in a facebook group how do you whitelist posts from certain users?" The answer is "you can't", but AI won't tell you.

I ask Gemini health questions non stop and never see it using YouTube as a source. Quickly looking over some recent chats :

- chat 1 : 2 sources are NIH. the other isnt youtube.

- chat 2 : PNAS, PUBMED, Cochrane, Frontiers, and PUBMED again several more times.

- chat 3 : 4 random web sites ive never heard of, no youtube

- chat 4 : a few random web sites and NIH, no youtube

To clear up potential confusion, you seem to be talking about the Gemini app (https://gemini.google.com), the chat app formerly known as Bard.

The article is about AI Overviews, a feature of Google Search (the LLM-generated box that sometimes shows up above the search results).

They're powered by the same pretrained model but, in true Google style, are two otherwise unrelated products built by two completely separate orgs.

Then there's AI Mode (https://www.google.com/ai), NotebookLM and probably some others I'm forgetting right now. :)

I would guess that they are doing this on purpose, because they control YouTube's servers and can cache content in that way. Less latency. And once people figure it out, it pushes more information into Google's control, as AI is preferring it, and people want their content used as reference.

Google AI cannot be trusted for medical adivice. It has killed before and it will kill again.

Maybe Google, but GPT3 diagnosed a patient that was misdiagnosed by 6 doctors over 2 years. To be fair, 1 out of those 6 doctors should have figured it out. The other 5 were out of their element. Doctor number 7 was married to me and got top 10 most likely diagnosis from GPT3.

With the general lack of scientific rigour, accountability, and totally borked incentive structure in academia, I'm really not sure if I'd trust whitepapers any more than I'd trust YouTube videos at this point.

The authoritative sources of medical information is debatable in general. Chatting with initial results to ask for a breakdown of sources with classified recommendations is a logical 2nd step for context.

Google AI Overviews put people at risk of harm with misleading health advice

https://news.ycombinator.com/item?id=46471527

the real promise of large language models is that before people were looking too closely, datasets had been procured in questionable ways. so then users can have access to medical data based on doctor's emails and word documents on their pc. which would have a lot of value. no it has become a glorified search engine.

It's crazy to me that somewhere along the way we lost physical media as a reference point. Journals and YouTube can be good sources of information, but unless heavily confined to high quality information current AI is not able to judge citation quality to come up with good recommendations. The synthesis of real world medical experience is often collated in medical textbooks and yet AI doesn't cite them nearly as much as it should.

The vast majority of journal articles are not available freely to the public. A second problem is that the business of scientific journals has destroyed itself by massive proliferation of lower quality journals with misleading names, slapdash peer review, and the crisis of quiet retractions.

There are actually a lot of freely available medical articles on PubMed. Agree about the proliferation of lower quality journals and articles necessitating manual restrictions on citations.

For straight up search Google is better, but for AI search I prefer Bing.

Ohhh, I would make one wild guess: in the upcoming llm world, the highest bidder will have a higher chance of appearing as a citation or suggestion! Welcome to gas town, so much productivity ahead!! For you and the high bidding players interested in taking advantage of you

Exactly. This is the holy grail of advertising. Seamless and undisclosed. That, and replacing vast amounts of labor, are some of the only uses that justify the level of investment in LLM AI.

What about the answers (regardless of the source)? Are they right or not?

imo, for health related stuff. or most of the general knowledge doesn't require latest info after 2023. the internal knowledge of LLM is so much better than the web search augmented one.

google search has been on a down slope for awhile, it's all been because they focused on maximizing profits over UX and quality.

Don't all real/respectable medical websites basically just say "Go talk to a real doctor, dummy."?

...and then there's WebMD, "oh you've had a cough since yesterday? It's probably terminal lung cancer."

WebMD is a real doctor, I guess. It's got an MD right in the name!

Google AI overviews are often bad, yes, but why is youtube as a source necessarily a bad thing? Are these researchers doctors? A close relative is a practicing surgeon and a professor in his field. He watches youtube videos of surgeries practically every day. Doctors from every field well understand that YT is a great way to share their work and discuss w/ others.

Before we get too worked up about the results, just look at the source. It's a SERP ranking aggregator (not linking to them to give them free marketing) that's analyzing only the domains, not the credibility of the content itself.

This report is a nothingburger.

> A close relative is a practicing surgeon and a professor in his field. He watches youtube videos of surgeries practically every day.

A professor in the field can probably go "ok this video is bullshit" a couple minutes in if it's wrong. They can identify a bad surgeon, a dangerous technique, or an edge case that may not be covered.

You and I cannot. Basically, the same problem the general public has with phishing, but even more devastating potential consequences.

The same can be said for average "medical sites" the Google search gives you anyway.

It's a lot easier for me to assess the Mayo Clinic's website being legitimate than an individual YouTuber's channel.

I don't think anyone is talking about "medical sites" but rather medical sites. Indeed "medical sites" are no better than unvetted youtube videos created by "experts".

That said, if (hypothetically) gemini were citing only videos posted by professional physicians or perhaps videos uploaded to the channel of a medical school that would be fine. The present situation is similar to an LLM generating lots of citations to vixra.

Your comment doesn't address my point. The same criticism applies to any medium.

The point is you can't say "an expert finds x useful in their field y" and expect it to always mean "any random idiot will find x useful in field y".

Imagine going onto youtube and finding a video of yourself being operated on lol

The assumption appears to be that the linked videos are less informative than "netdoktor" but that point is left unproven.

Randall Munroe coined the helpful word 'citogenesis' to describe the combination of uncritical citation and feedback loops. Once the nonsense is out there, it's hard to ever claw it back.

https://xkcd.com/978/

I've seen so many outright falsehoods in Google AI overviews that I've stopped reading them. They're either not willing to incur the cost or latency it would take to make them useful.

Probably because the majority of medical sites are paywalled.

I'm getting fucking sick of it. this bubble can go ahead and burst

Same energy as “lol you really used Wikipedia you dumba—“

How long will it be before somebody seeks to change AI answers by simply botting Youtube and/or Reddit?

Example: it is the official position of the Turkish government that the Armenian genocide [1] didn't happen.. It did. Yet for years they seemingly have spent resources to game Google rankings. Here's an article from 2015 [2]. I personally reported such government propaganda results in Google in 2024 and 2025.

Current LLMs really seem to come down to regurgitating Reddit, Wikipedia and, I guess for Germini, Youtube. How difficult would it be to create enough content to change an LLM's answers? I honestly don't know but I suspect for certain more niche topics this is going to be easier than people think.

And this is totally separate from the threat of the AI's owners deciding on what biases an AI should have. A notable example being Grok's sudden interest in promoting the myth of a "white genocide" in South AFrica [3].

Antivaxxer conspiracy theories have done well on Youtube (eg [4]). If Gemini weights heavily towards Youtube (as claimed) how do you defend against this sort of content resulting in bogus medical results and advice?

[1]: https://en.wikipedia.org/wiki/Armenian_genocide

[2]: https://www.vice.com/en/article/how-google-searches-are-prom...

[3]: https://www.theguardian.com/technology/2025/may/14/elon-musk...

[4]: https://misinforeview.hks.harvard.edu/article/where-conspira...

> Google AI Overviews cite YouTube more than any medical site for health queries

Whaaaa? No way /s

Like, do you people not understand the business model?

It’s slop all the way down. Garbage In Garbage Out.

Conflict of interest.

I believe we need to do something. I see the big corporations slowly turn more and more of the world wide web into their private variant.

> big corporations slowly turn more and more of the world wide web into their private variant.

Geocities was so far ahead of its time.

Google AI (owned by Meta) favoring YouTube (also owned by Meta) should be unsurprising.

> Google AI (owned by Meta) favoring YouTube (also owned by Meta) should be unsurprising.

...what?

This is absolute nonsense. Neither Google AI or YouTube are owned by Meta. What gave you the idea that they were?

Probably asked an llm

Sorry, I meant to type "Alphabet"

I imagine that it is rare for companies to not preferentially reference content on their own sites. Does anyone know of one? The opposite would be newsworthy. If you have an expectation that Google is somehow neutral with respect to search results, I wonder how you came by it.

How do I respond to this nicely without getting my comment flagged

People don't flag comments because of tone, they flag (and downvote) comments that violate the HN guidelines (https://news.ycombinator.com/newsguidelines.html). I skimmed your comment history and a ton of your recent comments violate a number of these guidelines.

Follow them and you should be able to comment without further issue. Hope this helps.

I feel like you completely missed the point of the rhetorical question.

It was a stupid question/post by me, I just don't know how to respond to "why should they not preference their own site? :)"

Because... that's not how it should work? And it makes something of a case for antitrust?

I do apologize, however with that being said this https://news.ycombinator.com/item?id=46767027 just got flagged

They do flag because of tone, or else outright things that don't fit with their agenda

(What I posted was very substantive)

Reply on news.ycombinator.com