---
title: 'Standards to Expect from YouTubers'
source: 'https://youtube.com/watch?v=xlM8X7KHveE'
video_id: 'xlM8X7KHveE'
date: 2026-08-17
duration_sec: 2387
channel: 'Adam Ragusea'
---

# Standards to Expect from YouTubers

> Source: [Standards to Expect from YouTubers](https://youtube.com/watch?v=xlM8X7KHveE)

## Summary

The video discusses the Hank Green AI controversy as a springboard for examining editorial standards and professional ethics in online fact-based media. The creator argues that while LLMs can be useful research tools, they should not be used for routine writing, and that the industry needs to develop and enforce professional norms, possibly through a trade association.

### Key Points

- **Hank Green AI Controversy** [00:13] — Hank Green was caught relying heavily on generative AI for research and possibly writing, which sparked a discussion about editorial standards in online media.
- **LLMs as Research Tools** [01:28] — Large language models have some utility as research tools, but they are not useful enough to justify the massive capital misallocation into speculative AI businesses.
- **LLMs as Secondary Sources** [04:17] — LLMs should only be used to find primary sources, not as a substitute for them. Accuracy requires going back to the original source to scrutinize reliability and give credit.
- **No AI for Routine Writing** [06:17] — Professional communicators should not use AI for routine writing. If you don't have strong feelings about what to say, you have nothing worth publishing.
- **Lack of Professionalization** [11:14] — The creator admits to never professionalizing his operation, despite being able to afford an editor, which could have caught mistakes before publication.
- **Wild West Period** [13:14] — Social video is in a pre-1929 stock market period of rapid growth and lawlessness, where old systems of professional ethics have collapsed and need to be rethought.
- **Need for Trade Association** [15:27] — The industry needs a trade association to draft and maintain a professional code of ethics, similar to other profitable industries.
- **Counterfeit News** [22:31] — The creator suggests calling fake news 'counterfeit news' because it looks like it has value but has none, especially when it's about the real world.
- **Trust as a Bank Account** [23:03] — Public trust is like a bank account: built up over time with good work, and debited when mistakes are made, proportionate to the perceived severity.
- **Comments as Self-Correction** [30:12] — Audience comments serve as a self-correcting mechanism in online media, similar to editorial review, though not always adequate.
- **Institution Building** [36:28] — Individuals can form associations to set standards, and Gen Z is encouraged to build institutions to address the moral correction needed in the industry.

### Conclusion

The video concludes that the online media industry is in a period of lawlessness, but individuals can choose to build institutions to set standards. The creator is too tired to lead, but encourages younger creators to agitate for higher standards.

## Transcript

So you've probably already heard Hank Green,  the very successful and well-regarded internet   science communicator, Hank Green got kind of  caught by his audience relying a little too  
heavily on generative AI to research and  possibly even write some of his videos.   I think this is a deal, but it's not a big deal,  especially considering how Hank has handled the  
fallout and considering the reservoir of goodwill  he has built with his audience over the years.   I do think this incident offers an opportunity to  examine how we are setting and enforcing editorial  
standards within the fact-based media business  online, especially in video form. What are the   professional standards for doing the kind of  work that Hank Green does, that I do? And how  
do we re-professionalize journalism and other  kinds of informational public communication as   those businesses have reconstituted here on social  video, or do we even want to re-professionalize  
those fields? That is the main thing that I have  written about for you today. But first, let's   talk specifically about ChatGPT and the like. I do think that large language models have some  
utility as research tools. They're not nearly  useful enough to justify the massive catastrophic   misallocation of capital into highly speculative  AI businesses that will never make enough money  
to return those investments. And the financial  crime, the mendacity and/or the negligence that   has inflated the AI bubble is a global scandal  that will hurt all of us, especially here in the  
United States, where we have already spent a huge  chunk of our national wealth on this last-ditch,   hair-brained scheme to save our country  from decline and eventual irrelevance,  
like the Darien scheme. Scotland, at the turn  of the 18th century, invested a fifth of its   national wealth to found an American colony in  what is now Panama, thinking correctly that the  
Isthmus of Darien, the narrow point between  the two American continents, that that would   become a crucial choke point for global trade.  They were right about that, but they were about  
200 years early, and they all died of malaria,  and Scotland had to be absorbed by England with   the Acts of Union to settle its debts. AI is the United States' Darien scheme,  
I fear, and large language models are not useful  enough to justify all of that, but they can be   useful for research. Sometimes I will wonder,  hey, have any scientists studied how cooking  
affects the pH of onions? And I'll go looking  in the research literature, Google Scholar,   et cetera, and sometimes I can't find anything.  So I'll ask an LLM that has scraped the entire  
internet, and sometimes the AI finds me a source  that I missed. "Yes, cooking does lower the pH   of onions," the chatbot will say to me, usually  with much more confidence than is warranted by the  
actual findings it is referencing, and sometimes  it outright mischaracterizes those findings,   and often it cites sources that I do not regard  as reliable. A post on Quora is not on the same  
level as a peer-reviewed journal article. I have found LLMs useful as a means to find   primary sources, but of course, that's really  all they should be used for when we're talking  
about researching the published record of all  human knowledge. The LLM is a secondary source.   It is at best a directory through which you can  find primary sources of information, which is  
the same lesson that everybody had to learn about  Wikipedia, and it's the same lesson they taught   us about Encyclopedia Britannica in the library  at Park Forest Elementary School, circa 1992.  
In situations where accuracy really counts,  you cannot rely on any secondary source.   You must use it to go back to the primary source  where the claim originated so that you can  
scrutinize the reliability of that scientist or  that eyewitness or whatever, and so you can give   them credit for learning what they learned, or  so you can make them accountable for being wrong  
instead of you being held accountable, and so that  you can prevent the inaccuracies that inevitably   result when you make a copy of a copy of a copy  of a copy. When a primary source is summarized by  
a secondary source, and then another source  summarizes the secondary source, and so on,   the game of telephone like that always scrambles  the details. All of these reasons and more are why  
we always go back to primary sources when accuracy  counts. And if someone is going back to primary   sources, then I don't care whether they get there  via Encyclopedia Britannica or Claude. Either is  
fine. That's one obvious ethical principle for  responsible LLM use that I am pretty sure about.   Here's another thing that I'm still kind of  kicking around in my head. I think it may be  
my position that a professional communicator  should not use AI for any routine writing   purposes. If you're a professional communicator  and you don't already have very strong feelings  
about what you want to say and how you want to  say it, then I don't think you have anything worth   publishing. No one needs you to read a chatbot  output for them. They can read it themselves,  
thank you very much. The only thing that I want to  hear from you is what you have to say. And indeed,   that is the only product of any potential value  that you have to offer. Anything else is just  
like slapping your brand onto a commodity.  That's great work if you can get it, but it's   not something that I would feel good about doing,  and it's certainly not anything I want to buy.  
If you don't have strong feelings about what  you want to say and how you want to say it,   then you probably don't have anything of value  to communicate, and asking the computer to write  
something for you does nothing to address  that core vacancy. I wrote the script that   I'm reading right now myself because I know a  lot about what I want to say. I wouldn't want  
anybody else to write it for me, let alone a  computer. Also, because the act of writing this   helped me to work through my own thoughts.  Writing equals thinking equals writing. If  
you outsource your writing to a computer,  you outsource your thinking to a computer.   So my provisional position on LLM use by people  in my general line of work is that it's fine if  
you treat it like any other research tool  to lead you to primary sources, but the LLM   ought not fabricate any substantial amount of the  stuff that you're selling, because that's like me  
slapping a vinegar leg logo on generic brand  jar sauce and then selling it back to you at   a steep markup. That's what I think, and I think  that's what Hank Green thinks. And may we all try  
to live our lives in accordance with our beliefs  a little bit more today than we did yesterday.   Anyway, I'm wandering all over the place today,  which is what you are able to do with Saily,  
sponsor of this video. Lauren and I drove over the  border to Quebec at the beginning of the summer,   and every time, we forget about the phone problem.  Oh wait, are we roaming now? That'll cost a  
fortune. Turn off your phone. But the phone is  giving us our directions. Should have had a Saily   eSIM. If you're traveling, you choose a Saily data  plan for any of more than 200 destinations they  
have, you activate the eSIM before the trip,  and when you land in a new country, your new   data plan activates automatically. No roaming. Saily has eSIM plans for people on cruise ships,  
hence the name. Set sail with Saily. That's  not their slogan. That's just what I'm saying.   Download the free Saily app. Choose the data  plan that's right for your business or for  
your travel plans, and be sure to save 15% with my  coupon code, adamragusea, or just hit my link in   the description, saily.com/adamragusea. And of  course, there's the QR code there too that you  
could scan. 15% off Saily data plans when you  tell them that I sent you. Thank you, Saily.   Anyway, the whole Hank Green thing got me  thinking about my own mistakes and my own  
little controversies as an ostensibly  responsible informational social video   creator. Does YouTube count as social video? I  think it's an edge case, but I'm going to put  
YouTube under the social video umbrella  for the purposes of this discussion today.   Many of the mistakes that I've made and that Hank  Green has made, many of these mistakes could have  
been caught before they went out the door. Many  of these mistakes could have been addressed   prior to publication by a more traditional,  professional editorial process, the kind of  
process that used to be standard operating  procedure in the fact-based media business.   Really, I think the principal sin of my YouTube  career is that I never really professionalized  
this operation, even though I probably could  have afforded to. I could have afforded to hire   a professional script editor, maybe some  food science grad student who also knows  
something about good writing. Such people exist.  I could have found one and paid them a salary.   And I maybe could have had another layer  of review with a freelance senior editor,  
somebody really seasoned who is in the consulting  phase of their career, perhaps after getting laid   off from a lofty position. Such people exist.  I know many such people. I could have paid  
such a person a flat fee to give me a final  edit on everything that I have published.   This would have cost a lot of the money that  I've made on YouTube, like maybe even half  
of the money, since I think I would need to pay  that technical editor quite a lot to retain them.   And if I did retain them, I think that they would  end up wanting to get into the production side,  
the video editing side, eventually maybe  being on camera, and that person would   probably end up my partner in this business.  And I don't want a partner for many reasons,   but maybe too bad, buddy. Maybe that's just  the cost of running a real grown up business.  
Here is the message in a bottle that I wish  to send to future media historians. Some of   us who are living through this early boom cycle  of social video right now, some of us know that  
we're living in a Wild West period, a pre-1929  stock market period, a period of really rapid   business growth with a tumorous gold rush attached  to it, and a period of vast, effective lawlessness  
and normslessness. We're in the period where a  technology-driven disruption of the mass media   business has brought about the collapse  of old systems of professional ethics,  
which legitimately need to be rethought and  rewritten for this new world that we're in.   But until we do that en masse as a community of  professional media creators working on social  
video, until we come together to develop and  enforce new updated professional norms, until   we respond to some collective crisis by pledging  to endorse and to abide by a written code of  
professional ethics, until Mr. Beast finally does  something so bad that costs him very dearly with   his audience that he needs to be able to publicly  endorse and commit himself to some written  
professional code of ethics, until that particular  exigency or one like it arises, we probably won't   come together to build a professional community  institution that would draft such a code of  
ethics, and put it to a vote by its members,  and assign a standing ethics committee to   review the ethics code with various layers and  phases of community input, and to revise that  
ethics document every two years or something. Until we as YouTubers, as aspiringly responsible   social video makers of all kinds, until we form  a trade association for the same reasons that  
most other profitable industries have formed  trade associations before us, until we do that,   probably in response to some big crisis in  trust, until we build that institution to draft  
and maintain a professional code of ethics among  other things to serve our professional community,   until we found the International Association  of Content Creators, and there is at least one  
effort underway to create such a thing right  now, until we professionalize like that, our   industry is in a period of relative lawlessness  and/or normslessness. And in such periods,  
you generally see some level of disinhibition. We can all get away with so much right now   because the rules are still hazy at best, and bad  actors are all around us and among us, setting  
the ethical floor for the social video business  squarely on the seventh level of hell. I mean,   how much more cynical and irresponsible could you  get than the worst engagement bait content that's  
flooding the internet right now? Because  that is our current business environment,   it's easy for somebody like me to do the easier  thing, which is to run a six-figure annual media  
business without employing or contracting  with a single professional editor. Mind you,   I'm not talking about a video footage editor.  I'm talking about a content editor, a person  
who reviews the content for accuracy, for clarity  and perhaps style, an editor who stakes their own   professional reputation on the quality and the  responsibility of the content that they edit.  
Most people who used to make a living  writing or talking seriously about or   photographing or otherwise documenting with  media stuff that is happening in the world,  
most such people who did that for a living one  single generation ago, most of them usually had   professional editors overseeing and signing off on  their work. A lot of us don't have that right now,  
many because they can't afford an editor, some  of us because we're enjoying being relatively   early to a raging party with the parents out  of town for some undetermined amount of time.  
A lot of us are just still so happy to be free to  write and say what we want without having to get   daddy's permission, we're still so thrilled to be  out on our own making it, making it kind of big,  
all without grownup supervision. At some point in this growth cycle,   you have to admit that you have become the  grownups, and standards will only exist if they  
are upheld by the grownups, because there's no  one else to do it. That's what we call growing up,   when you realize that there's no longer somebody  else in the group in charge. It's either you and  
your group who's in charge or no one's in charge.  We haven't matured that far as a community of   responsible content creators. I haven't matured  that far individually in the work that I do here.  
It should be said that some content creators  have long explored these frontiers of   professionalization or re-professionalization  of the media business. The Green brothers  
have made so many quality responsible videos in  collaboration with named, credited, professional   content editors. And then, they've made some  other things without as much editorial review.  
Hank Green made a thing or a few things  that were not professionally up to snuff,   and it's logical to speculate that these missteps  could have been avoided or minimized with more  
layers of professional editorial review,  like we used to have almost uniformly in   the fact-based media business because the  fact-based media business runs on trust.  
You don't have to trust all the content creators  whose work you consume. Sometimes you just want   to sit around and watch an amorally violent  action movie, and sometimes you want to sit  
down and scroll shameless cynical brain rot on  vertical social video. You don't need to trust   the people who make that junk for you. But if  you are consuming content where the people are  
talking about real things happening in the real  world, you need to have some level of trust in   that person who is talking or else you're not  going to listen to them. If they're just making  
stuff up to maximize engagement, that has no value  to you, so you probably won't consume it twice.   In the world of media scholarship, we used to  call such content fake news. Not slanted news,  
not biased news, not propagandistic news, but  fake news, claims that people literally just   make up and flood the internet with as part  of some kind of low-margin cash grab scheme.  
That's what we used to mean by fake news. But  then, in 2016, Donald Trump appropriated the   term to describe news coverage that  he perceived as unfavorable to him,  
and the term became useless to media scholars.  I suggested calling it counterfeit news instead.   Counterfeit, a product that is made to look like  it has value, but in fact has no value at all,  
because if a piece of media that's about the real  world isn't real, then there's nothing for you to   learn from it, and so it is useless to you. This is a trust-based business that I'm in,  
that Hank Green is in, that a lot of us are in,  and public trust is not binary. It's not something   that you either have or don't have. You can have  some trust, but not a lot. You can have a lot of  
trust, but then lose a little, et cetera. A public  figure or prominent professional builds an account   of trust with the public over time by doing  quality, responsible, dependable work. The more  
good work you do, the more you prove yourself, the  more trust that you accrue in your trust account.   And when you do something that undermines  the public's trust in you, as I have done,  
as I'm pretty sure everyone has done at  least once to some degree, all have sinned,   when you make a mistake of craft or a mistake  of taste or a mistake of professional ethics  
or perhaps even a moral mistake, when you  shake that bond of trust with your audience,   your balance of trust in the bank goes down. But  it goes down in an amount roughly proportionate  
to the public's perception of the severity of  your offense against them, and that outflow of   trust is debited against the balance of  trust that you have previously built up,  
which may be a considerable sum, perhaps  a sum that dwarves the debt incurred.   I think that is what generally happens and  what should happen and what will happen when  
a public figure errs, as we all have to widely  varying degrees. My point being, some have tried,  
are trying, will probably try soon to do more  professional work on the internet. And I'm   not going to attempt to define professional media  work here, but I think that one obvious element of  
professionalism in fact-based media is some layer  of editorial review. That is, I'm pretty sure, one  
of some debatable, hazy cloud of attributes that  would naturally define professionalism in media.   There are some trying to reprofessionalize the  fact-based media business that lives here on these  
internet shores, where some lucky few of us washed  up and colonized and thrived with some combination   of skill, moxie, dumb luck, and cynicism. That  fact-based, trust-based media business that lives  
on in some mutated, evolutionarily-bottlenecked  state here on the internet. There are some people   who've tried or are trying to professionalize  that media business, which I happen to be  
in. But I'm not really one of them, am I? If I think that professional, supervisory,   and to some extent, adversarial editing from  a professional editor, if I think that is some  
basic element of a professional code of ethics  that should exist and may one day exist as a   commitment that one must make, at least implicitly  for access to some segment of the media market,  
if I think getting an edit is a necessary  condition of professionalism in the fact-based   media business, then why don't I hire an editor?  I could afford one. I couldn't pay as much as  
I maybe could have justified during my peak  earning years. COVID, basically. I couldn't   pay as much anymore, but I could still  pay a fair market rate to a professional   editor on a consulting basis, freelance. But I don't because I really don't want to,  
because I'm still enjoying being free, because I  don't work well with other people, at least not   on the project level, because I work odd hours  and I usually work right up against my deadline,  
in part so that my channel can be  more responsive to current events.   And yeah, I don't want to hire an editor because  I don't want to share the money. Even though I do  
have more than I need right now, I'm too worried  that it's all going to collapse one day, and I'm   still kind of hoarding treasure to bury and dig  up if and when my fortunes inevitably reverse.  
And right now, I can get away with not having an  editor because my audience does not demand it, and   I'm not a member of any professional association  that recommends it, and the legal system has not  
yet prompted me to institute an editorial review  process, which would help me to demonstrate to   the court that I did my due diligence in,  for example, a defamation suit against me.  
That's part of the reason why the old fact-based  media had editors, because they reported   unflattering things about real people, and they  occasionally got things wrong. And you're more  
likely to lose a defamation suit against you  if it seems like you've been reckless with the   truth. If you can show that you have editors who  are at least theoretically enforcing some written  
code of ethics, even if that process occasionally  produces an inaccurate result, if you can show   that you have a process, that can improve your  position in a legal dispute about your content.  
That hasn't happened to the likes of me  yet on the internet, probably because   we online-content-creating micro celebrities  represent a wide field of small legal targets  
as opposed to one big, juicy legal target that you  can sue, like The New York Times or whatever. That   practical legal reality is probably another reason  why someone like me hasn't professionalized,  
and why someone like Hank Green has kind  of semi-professionalized. Some of the work   he publishes has serious editorial review,  and some of it seems to be stuff that he's  
just kind of dumping from his brain straight  to the internet, the way that I usually do.   And there's one more reason. I, for one, still  wonder if we will actually need as much editorial  
supervision in new media as we had in old media,  because editorial supervision is not an end unto   itself. The function of editorial review is to be  a self-correcting mechanism within the fact-based  
media business. That function can be performed  in different ways by different people. It may   not be adequate, but the comments do represent  a self-correcting mechanism in the online media  
business. Rapid, voluminous, public audience  feedback has demonstrated some efficacy as a   quality control instrument in new media. When  I'm out there on the internet and I see someone  
making a claim that I'm pretty sure is hogwash,  there's usually at least one or two top comments   pushing back on it, including on my own videos. Rideshare drivers don't need a professional  
association to enforce standards within their  industry. They need a union to represent their   interests in disputes with rideshare companies  who are profiting handsomely off of their labor,  
but rideshare drivers do not need a professional  association to set and enforce norms because the   rating system takes care of that. Good drivers  get good ratings from their riders and bad drivers  
get bad ratings, and you can see the rating when  you choose to accept a driver or not. It's not a   perfect system. It can be abused. But so far,  it seems to be working pretty well, I think.  
We don't need editors for the sake of having  editors. We need accountability. And the internet   does bring some measure of accountability to the  people like me and Hank Green who do business  
here, through comments and Reddit threads and the  controversy sections in our Wikipedia bios, all of   that. The question is whether that accountability  is enough, and how do you define enough? Enough  
for what? Well, enough accountability to maintain  the level of trust necessary to have a successful   fact-based media business on the internet. That  is the cold economic answer to the question, and  
I guess we're going to see what the answer is. For myself, I'm not ready to commit to hiring a   script editor to review the accuracy and taste  and style of all my work here before I publish  
it. Maybe one day, something will happen where  you demand that of me, but that is not today,   probably. I'm semi-retired. I'm old  and tired. Overtired for my age,  
honestly. Maybe it's not the years, honey, it's  the mileage. And I'm honestly not interested in   doing my best work most of the time anymore.  I don't want to half-ass anything, but I do  
occasionally quarter-ass some jobs. Sometimes I  just tell you what I suspect to be true instead   of calling up a bunch of experts and telling  you what they think is true, which would be a  
more valuable service for me to provide to you. Frequently, these days, I take the shortcut and   I just tell you what I reckon is going on with  some news story or with some science question,  
which is a debit to my trust account, but I  make a little bit back by being transparent   about it. I'm honest and upfront about the fact  that I'm old and tired and a little lazy at this  
point in my career, at least when it comes to  relatively inconsequential topic areas. I'm   very careful to cite my sources, even when the  source is just my own intuition, especially when  
the source is just my own intuition. I make an effort to tell you that,   to explicitly distinguish between things that I  think are true and things that are proven true,   or things that are widely believed by experts  to be true, or things that are believed by some  
particular expert to be true, but not by others,  et cetera. I'm pretty good, I think, about   explaining my epistemology as I go, and I think  that earns me back a little bit of the trust that  
I lose, so that when I quarter-ass a job, it's  effectively like I'm one-eighth-assing the job on  
the adjusted trust index, and that is apparently  good enough for you. You're still here for now.  
We'll see what levels of professionalism  are necessary to thrive in this market in   the coming years. And of course, we need not look  at this through the dehumanizing economic lens,  
where we're all just rats in a changing maze  responding to the dynamics of the maze according   to broadly predictable behavior patterns. That  is the economic lens, if I were to describe  
it uncharitably. If we look at this through a  humanistic lens, we can say that people have free   will, people have power, and they can choose to  exercise that power to try to get what they want.  
If you want more better fact-based content on the  internet, regardless of what that market seems to   want economically, well then form an association.  Yes, institutions and norms are collapsing all  
around you right now, that's the scary part.  But the exciting part is that you get to try   to rebuild. Maybe Gen Z will be great institution  builders, like the World War generation was. Maybe  
Gen Z's annoying incessant moralism is a very  natural response to the amorality and the outright   immorality that they see being practiced and  rewarded at the highest levels of society today.  
It is past time for a moral correction. Maybe  that is the task that Gen Z has inherited.   Moral corrections generally require institution  building, and you don't have to ask anybody's  
permission to build institutions. Do you have  two like-minded friends? Well, congratulations.   You have already formed the society for  the advancement of whatever it is that  
you three care about. Maybe think about  making it a legal entity at some point,   like a nonprofit or something. If you can't  figure out how to do that, you can hire a lawyer.   They usually help you to incorporate for cheap,  like as a loss-leader for future business that  
they may get out of you for being counsel to the  society for the advancement of whatever it is.   There have been efforts to organize online  video content creators before. There are  
efforts underway now, and hopefully one of them  will catch at some point. At the very least,   we need a professional association to  represent our interests with the platforms,  
with the YouTubes and the TikToks that we are  making rich without having a clear and fair   system of rights and expectations to mediate  disputes between creators and platforms.  
I'm too old and tired, but you young people should  choose to participate in that institution-building   process, and if you want the creators that  you consume to be held to a higher standard,  
agitate for it. Institutions are usually formed to  placate the mob, so make yourself hard to placate,  
and you'll probably get more of what you want  from me, from Hank Green, from politicians,   from business execs, from anybody. Squeaky  wheel gets the grease. Talk to you next time.