Why We're Banning AI
currentaffairs.org
currentaffairs.org
Unemployment in hundreds of millions globally? Loss of part that makes us humans in many creative aspects? People getting visibly and measurably dumber and lazier, including top performers?
Absolutely 0 solutions, nobody cares. People here on HN often just think its not their problem, they can cash those bonuses and rest of mankind can go fuck themselves and serve them for near nothing, being on top /near top of the pile is the ultimate goal. At least this seems to be unspoken part of many llm discussions here. Pathetic. It seems this is way more typical mindset in US compared to say in Europe and tells a few things about current US society, but nothing nice.
We are losing a significant part of our humanity if llms really deliver on their promise and not just crash global economy big time which on its own is huge problem. People steering this are mostly typical high functioning sociopaths like altman focused on money and power, so dont expect those to care.
So yeah, there are concerns to say the least, not everybody operates on autistic narrow tube view of the world as themselves and thats it. Some of us have kids too.
Making a declaration that you are rejecting AI can come across as implying use of AI is a moral failing. There are certainly people who overtly claim it is a moral failing.
I think the people who are offended are perceiving this as as statement impugning their character.
The reasoning you describe seems somewhat self centered.
I'm not sure I want to worry too much about people who go through life assuming everyone else's statements are a direct commentary about themselves.
Would you think it reasonable for a mens health magazine to choose to not feature obese people on the cover?
But on the other hand,
Would you think it reasonable for people to feel singled out when they choose to make a public declaration that they will never put an obese person on the cover?
Being obese or saying "thin only" is not the same as "no ai content".
For producing content... yes. yes it is.
How dare you expect to vomit into the street and expect me to treat it like flowers.
This trend in other areas of life is becoming boring and tiring.
The double think is astounding.
They are offended by your vitriol and the tone that suggests the claim is indisputable.
There is no double think there.
There's nothing moral about using AI to replace the same artists whose work was used to train it.
These folks will need some serious deprogramming in a few years.
Sure, AI can generate slop at a very high rate, but it can also help writers reformulate a text or act as a proofreader. The editor's choices matter, as do the result and the audience's taste.
However, lately, there has been growing AI fatigue, and people are probably becoming willing to pay more if they can be sure that there's "No AI inside" (something like "No palm oil inside"). If people are willing to pay for it, we'll see.
Personally, on my blog, which I moved away from Blogger and which has been deindexed by Google, I have the same statement: 100% handwritten (with a disclaimer that I am not responsable for crawler hallucination). Why? First, because writing is really the fun part. I make better editorial choices when I write the text directly. Second, there's the audience. On my blog, over 90% of the audience consists of bots and crawlers. They need handwritten stuff, and if this audience grows, it probably means they like those posts.
In the end, the funny part is analyzing my text with LLMs after a post is published. The result is surprises and more ideas. If I ever ask what I need to change to make the text sound more human, the response is usually that I have to add more slop. That's the part I leave out.
It’s not about being better or worse.
Anyone can sling LLM generated text around, it’s boring and uninteresting.
I'm curious about other possibilities, like how we can incorporate AI to produce higher quality content, more quickly. Or more accurately. Bosses at work seem to assume that the way AI would benefit us to speed things up by some integer factor, and I kept explaining that tools like AI can provide benefits that don't appear in the "speed" dimension, per se.
One example is we can now vibe-out multiple potential ways for a feature to work, and actually play with them all over the course of an afternoon to help decide which one works best, and then actually code that particular approach for production.
I think the piece would be stronger if it explicitly explored ways to use AI in the authoring process that brought other benefits. As they note near the end of the article:
> It’s not going to be easy, because competitor publications are going to be able to produce immense quantities of writing in the time we are painstakingly correcting and adjusting a single article.
Right, it's fairly easy to generate a bunch of crap with AI. But what about having the AI critique your first drafts, or find flaws in the logic presented in an argument. I hope that writers work with AI in a sort of "rubber duck"/"sounding board" fashion, but it's hard to tell how they are actually working. The current front seems to be a hot debate about whether text was AI generated, but there's a lot to AI methodology beyond just that.
AI generated text is still exhausting to read.
It's entirely possible that people have just finely tuned AI to perfectly mimic them, but if it passes the Turing test, I guess I have nothing to complain about. Our rules are less about some fundamental opposition to AI and more that it's just annoying to read unedited AI slop.
Not that I would ascribe actual intelligence to "AI", but those fans who claim it's better than your average human almost seem to be proven right by the sheer amount of incredibly stupid behavior around "AI".
I've never seen AI writing be totally undetectable or on its own particularly good, but if it someday gets good enough that AI writing is indistinguishable, I'd still appreciate when someone goes out of their way not to use it.
What do you think is a more artful way of living, spitting into food undetected, or respecting the choices and boundaries of other people?
I think the far bigger and more important issue is that through generative AI hallucinations and lies become more and more part of our written media.
Albertatech told a humorous, light-hearted story[1] about how this affected her personally, but the results could be far more devastating:
The false AI generated content getting a lot of milage indeed is the big problem. So many fake AI slop shorts on YouTube for example.
Outside of a few standouts, the companies that survived did not do so by embracing quality journalism and independent reporting.
Now the bloggers and upstart publishers that killed traditional media are the incumbents.
Out of curiosity, which companies embraced quality journalism and failed because of it?
More importantly, you can add in all of the local independent newspaper, tv, and radio channels that were either shut down or gobbled up by large corporations and gutted (Sinclair, Clear Channel, etc). Over 3000 newspapers were closed down, as well as nearly every foreign or government bureau.
This is when the largest drop ever of professional reporters in the US happened - we entered the 2020s with less than half of the number of employed reporters as we had in 2000.
The slopification of news started before AI.
"Out of curiosity, which companies embraced quality journalism and failed because of it?"
Advertising revenue stolen by so-called "tech" companies acting as intermediaries (middlemen) had nothing to do with staff reductions, i.e., "failure"
Is that what this reply is suggesting
What if the success or failure of news organisations isn't necessarily a result of the quality of the journalism they produce
What if there is some other factor
For example, control over distribution (losing it to middlemen)
>This doesn’t mean nobody doing work here can ever use any of the softwares that are broadly lumped together as AI. Much as we respect the Luddites around here, we’re not against technological innovation. Nothing in our rules prohibits writers from using a new tool to help them research something, or prevents an artist from using a tool to, say, add a texture or change a color, or make some kind of small element.
As someone trying to pickup writing fiction after years of doing technical writing, I use LLMs for fast feedback to see if my words are having the intended effect (i.e., are they inducing emotions? am I describing things in interesting ways? or is it a slog to read? etc...). Since Current Affairs is a commentary magazine and not just straight facts reporting, I could imagine authors wanting similar feedback. Would this kind of usage be permitted? Seems... yes? But I'm not completely sure.
You trust LLMs to play tastemaker?
AI generated writing is (usually) hot garbage.
How would an LLM tell you this with any insight that is greater than your own?
Why? Your chatbot has ha no more experience of emotions than your teapot.
AI is great for some things. It is not currently great (or even remotely passable) at writing interesting magazine articles.
> Nothing in our rules prohibits writers from using a new tool to help them research something.
Why?
Well they didn't do that, and in fact specifically mentioned how they are not doing that. So what are you even talking about?
I can appreciate clever writing by humans because another human did it. Maybe I'd still critique it for being convoluted. But there'd be some kind of interaction with me and another person, even though I might leave with a negative opinion/feeling.
If an LLM tries clever writing, it's just the result of a statistical model using its dataset. There's no joy in reading the extra fluff because I know there's no-one that spent the time on doing it. In fact, I know that the person on the other side didn't want to spent the time doing it. And it leaves me wondering why I should even read all of that in the first place. If we're using AI to be more efficient, why not have it just generate boring, efficient, articles that just give you the information without pretending to present any opinions or jokes.
TL;DR if you want to be efficient, be efficient; don't generate fluff
Correct me if I misunderstand your implication, but it seems to be "I have seen poorly-written articles with obvious LLM tells and factual mistakes, therefore LLMs are incapable of writing good articles". It's the same logic people use to say things like "criminals are so stupid, just look at these prisoners", not realizing that their sample selects for incompetence.
There's no incentive whatsoever for any institution to credit an LLM for an excellent article, and in fact there are many incentives to avoid disclosure of any AI use. Typing "write me an article about $TOPIC, make no mistakes" into ChatGPT yields quite different results than the LLM-driven content generation pipelines major institutions have already been using for well over a year now.
This was so well-written and convincing that I just subscribed. I believe they are doing this the right way, and more power to them!
Obviously the WSJ staff shouldn't be using it in the final output, but if it's a prominent guest's opinion filtered through ChatGPT, I'm mostly fine with it.
It was an op-ed, it's inappropriate to edit those heavily. So they either had to accept it more or less as submitted (with the usual basic editing around grammar, clarity, and typos), or not accept it. If they'd rewritten it, it would not have been an op-ed by Druckenmiller anymore, but someone else's interpretation of his submission.
Maybe that would have been a totally acceptable outcome?
Missing the opportunity to be informed about those points because of aiphobia seems suboptimal.
ChatGPT polish adds no value to this.
AI used to be obvious, but the quality has improved — in 6 months it will be impossible to detect.
Eventually you will have to control the entire supply chain, like Dune. You will have to have an entire org policed with anti-AI software, attestation and spot checks to attest that AI was never used in the workflow.
The auditors will have to spot-check every employee randomly the way USADA / IADA (anti doping agencies) do – on-call within 2 hours notice for an anti-AI attestation review.
1 drop of AI taints the entire institution.
(I’m not anti AI, I’m explaining the implications. And I think the principle is good for their journal, though I don’t think they’ve thought through how to manage it yet. )
Small loops keep social trust high. When loops need to trade, they establish simple bilateral treaties with each other. This preserves value, protects human labor, and keeps decisions in local hands.
[1]https://www.makonea.com/en-US/casual/pangram-doesn-t-seem-ri...
I would way rather read handwritten articles. I think everyone should celebrate the human craft of writing. I really hope this doesn't slowly become the "no-VFX" posturing from the film industry, however.
So "AI" tool OK?
> to, say, add a texture or change a color, or make some kind of small element.
So "AI" OK for elements of some value of "smaller"?
This ban sounds somewhat ... flexible.
We shouldn’t give the AI companies even more avenue to influence people, some things must remain human.
if OTOH Fox News came out with this announcement, nobody would care except some of their viewers/readers would be annoyed by it.
point is it's hard to ever identify if a magazine would be doing this purely out of principle since it's also good marketing.
Humans are the new cockroaches. Our exorbitant and decadent carbon footprint not so much.
We are witnessing the natural outcome of pursuing higher profits at the expense of everything else; at some point it becomes “as crappy as the market will bear”, and we all lose out. For some of us it may manifest as having to get our car serviced every few thousand miles because the parts are of poor quality (but cheap!). For some of us it may manifest as no longer desiring sweets because they’re now poorly made with substandard ingredients created in a lab somewhere.
It can also manifest in ways that are a threat to public safety; like rollercoaster rides that weren’t properly designed or tested because schools now pursue profits above learning; or because manufacturers somehow rely on regulators to do their job for them.
However it manifests, it’s hard to point to higher prices and lower quality goods and services and say this is the best that capitalism has to offer. If this is it, if this is the capitalist system working as intended, it’s no wonder movements like the Democratic Socialists of America are taking hold.
This is, I think, as close to ideal as an AI policy for media can possibly be. For certain things - visual art, music, storywriting, game design, and public communications - I think consumers will continue to value the human element and demand it. The optimistic take is that this will line up with the capitalist incentive as well, no matter how much AI reduces the cost of production to nearly zero. People just won't want to engage with slop.
>Nothing in our rules prohibits writers from using a new tool to help them research something, or prevents an artist from using a tool to, say, add a texture or change a color, or make some kind of small element. But we need, first, to ensure that technologies are a tool used by workers to help them, rather than a tool used to replace workers. And, second, we need to make sure that readers aren’t being fed things that look like people’s genuine thoughts and creative expression but are in fact the product of disembodied mathematical formulas.
This is another W as far as their AI policy goes. While skeptical of "slop" and other kinds of lazily-generated, low-quality output, I also remain highly critical of anti-AI absolutism. Generally these are the types of arguments you see bandied about on places like Bluesky which lean hard on the environmental impact/third-worldist perspectives, argue for AI-driven skills loss, make tenuous and contrived connections of AI with their hot-button cultural issue, or only seem to view AI through the lens of their own job security. None of these issues are inherent to AI or are insurmountable, and hardliners make an incorrect assumption that the technology can simply be banned or boycotted away. Many of them are feverishly waiting for "the bubble pop", which is barked with the same optimism as evangelical predictions of the Rapture. Some of these people are so obsessive about being anti-AI that they wield whatever diminutive influence they may have to ban your IP address from their website if their server ever catches Claude doing a web search on it.
I think this is key. If AI wasn't force-fed into every industry in a rapacious grasp for more profits, our society might have been able to grow into a healthier equilibrium where AI is a useful tool for creative professionals.
Even the insiders I talk to at big AI labs express the opinion, "nobody likes where we're going, but the pressures are too strong."
Good luck trying to figure out whether something was written by AI or a human. This mass hysteria around AI will soon be forgotten. Nobody will care whether something was written by an LLM or a human. People will just move on to other topics.
Downvoting me or others will not change this fact.
It’s a magazine, not a social media website. They control who writes for them and they know who they can trust. If any of those people break that trust, they are banned and replaced with someone else.
> Nobody will care whether something was written by an LLM or a human.
Of course they will. To this day people still care about (for example) colonialist actions from hundreds of years ago or buying print books. It’s disconnected from reality to believe everybody will simply not care who wrote what they read.
Nor have I claimed such a thing. I recommend you read the rest of the paragraph instead of stopping at the first sentence and making a bad faith argument. If you don’t read a point to the end, you can’t understand it enough to argue against it.
That is not a bulletproof system. Human trust is not deterministic, it's not precise, and it's based on biases. That's the point.
> If any of those people break that trust, they are banned and replaced with someone else.
Do you understand what that means? I’ll help again. I means that I never claimed it was a “bulletproof system”, as you keep insisting. In fact, that makes it pretty explicit that I know it’s not bulletproof, because I exposed a failure case!
Here’s the thing, though: It doesn’t have to be bulletproof, and it is absurd to believe it does. It just has to be good enough, and you can get good enough by trusting people and evaluating on a case-by-case basis. Because you know what? Not everyone is an asshole who wants to fool others. The only people who believe everyone else is untrustworthy are the people who are untrustworthy themselves.
Btw it is kind of weird how it still can't write prose normally but it can write impressively high quality code now. Must be something to do with their training. Or maybe it's easier to quantify good code.
I don't understand your point. They're getting better so quickly it's hard to keep up.
Development and improvement is often very non-linear, assuming you can extrapolate forever is just as much guesswork as claiming it'll never change.
Just like they've forgotten about tulip bulbs?
Even if LLMs deliver a fraction of what their advocates promise, their impact entry into society will be relevant for centuries, much as with piracy, industrialization, and abolition.
Figuring out whether something is written by AI or a human is fairly easy with Pangram, and soon all major models will embed fingerprinting and it will be basically 100% detectable.
> Nobody will care whether something was written by an LLM or a human. People will just move on to other topics.
Do you have a crystal ball that tells you this, or is it just your gut feeling? Last I checked, AI and AI-generated content continues to be deeply unpopular outside of insular tech circles.
The value of written communication is in whether people decide to read it. There is nothing controversial about readers rejecting content they perceive is low-effort.
This seems to be the underlying moral justification for creatives' opposition of AI. Job destruction is the obvious self-interested justification.
My question (probably not original) is this: which biological neural network was not trained on copyrighted material from other artists? Which guitarist never listened to classic rock bands? Which painter never looked at Monet? Nearly all successful artists publicly recognize the influence of prior art on their development and style (and the rest are lying).
A quick hypothetical:
Let's say instead of hoovering up billions of images from data dumps, the AI researchers took a still camera to all the art galleries on the planet and snapped photos of art, then trained the neural network on those photos. Is this still "theft" of the artists' work? What about if an embodied AI in a robot chassis did the same, then painted new art with oil & canvas? Is this still theft?
Perhaps this example fails because it's not enough data? But is it a difference in degree or a difference in kind?
Those successful artists accessed their training data through legitimate means (please hold your 'well actually' replies about some artist using bittorrent here and there). They didn't all pirate, on a historic, industrial scale, every scrap of creative work that they saw in their lives. That's the key distinction.
It's important to frame these discussions in human terms. "Intellectual property" is a human invention and the rights associated with it are the results of negotiations between concerned parties. There was no negotiation with the training of the models, though. So no transaction happened, one group is profiting handsomely and whole classes of society which involuntarily had their labor taken are now being harmed.
There's no need to get it twisted, it's frankly quite black and white. The only question is how we'll ensure that the economics are corrected going forward to prevent further harm.
The premise that rules should equally apply here is absurd. It’s completely natural that certain rules don’t apply to humans but not AI and vice versa. Similar to humans vs animals or humans vs inanimate objects. Even among humans, rules don’t apply equally always.
Diffusion models tune a series of numbers via a complicated but essentially straightforward and "pipeline-like" process from the training data, so their work is significantly less derivative, even if visually it looks different. I guess we all agree that say, a encrypted zip file of a song is not a derivative, it's the song after some data processing.
Maybe we should improve the definition of derivative work to express how many layers of indirection lie between training material and output? If I trace over a painting I guess most would agree I've barely done any art, and I think that the process is more or less "pipeline-like" (even if complicated): input data is immediately converted to "art" by my brain following the lines. An artist that studies Goya for years and slowly improves his Goya-like portraits is another kind of level of indirection between training material and output!
Just go look at $_SECTION in your local bookstore. Most of the books have covers that fit right in with all the other covers on the shelf.
So my question is whether a computer doing the same thing is a difference in degree only, or in kind? But I think most people with strong opinions (and itchy downvote fingers) are coming at this from a moral outrage perspective (massive theft of IP / disregard for humans).
Nobody cares if hobbyists make a "Star Wars but make it Monet" art piece with AI and put it on their wall. It's the part where bosses fire John Smith and then ask the LLM, "please write 100 articles in the style of John Smith." It's the part where an actor is hired once, their face scanned, and then their likeness is used (unpaid) in a hundred films.
What about other artifacts that you produce, such as images (that are not illustrations), videos, sound effects for your podcast, or the code that runs your website?
> No AI-generated art may be used as a Current Affairs illustration
> am proud to say that as part of our new union contract, Current Affairs recently adopted one of the strictest anti-AI policies in the publishing industry. No text that is AI-generated may be published in our magazine. No AI-generated art may be used as a Current Affairs illustration.
Even if you want to keep AI out of the content-writing part, or the image generation part of your product, there are still areas where a newsroom can do much more ambitious work with the same people:
- AI can do research to assist in fact checking of articles
- AI can help suggest story ideas based on patterns it found in your own articles over time (an editor would still curate them)
- AI can help the newsroom pore over more first-hand sources/raw sources. Think of how many times you've had to process a stack of hard to OCR PDFs that come out of a FOIA, for example, or poring over government datasets when trying to make sense of something. Or a person's historical tweets, etc.
So it's clear that whoever won in this debate (hopefully there was at least a proper debate) wasn't interested in talking about all these additional areas where AI can help the existing people do more _without risking their jobs_.
They banned AI-generated text or images from appearing in the magazine.