I commented some version of the stated theory when the original article was posted, and having read this post, I don't agree with the idea that "the same basic pattern applies to basically anyone who is trying to make money by doing anything", at least not the examples given.
Pizza restaurants want to sell pizzas and are incentivized to produce quality, affordable pizza so people come back regularly. Customers want pizza (every now and then) and will visit pizzerias that offer tasty pizza, made with quality ingredients, in a pleasant setting, at a reasonable price. Ergo, incentives are aligned.
Dating app users (generally) want to find a partner and leave the app, ideally never to come back. Dating apps make money when users fail to find partners and pay for features that "increase their chances". Now, incentives are unaligned.
I run a semi-popular open source project (https://romm.app/), and this is a topic we tend to revisit regularly. While there will always have to be someone at the top who owns the project, we've tried to organize ourselves in a way that should prevent a complete hostile takeover:
* Gihub organization is co-owned (2 Owners)
* I own the domain, they run the Discord server
* Finances are handled by https://opencollective.com/
* All code is GPL or AGPL licensed
In the event either (or both) of us step away, temporarily or permanently, the core team is has the power and permissions to continue running the project indefinitely. While I would be able to remove them as co-owner on Github in a takeover scenario, I won't have access to the finances or the Discord community.
Because AI gets things wrong, often, in ways that can be very difficult to catch. By their very nature LLMs write text that sounds plausible enough to bypass manual review (see https://daniel.haxx.se/blog/2025/07/14/death-by-a-thousand-s...), so some find it best to avoid using it at all when writing documentation.
But all those "it's AI posts" are about the prose and "style", not the actual content. So even if (and that is a big if) the text was written using the help of AI (and there are many valid reasons to use it, e.g. if you're not a native speaker) that does not mean the content was written from AI and thus contains AI mistakes.
If it was so obviously written by AI then finding those mistakes should be easy?
The style is the easiest thing to catch for people; GP has said that the technical issues can be more difficult to find, especially in longer texts; there are times where it indeed are caught.
Passing even correct information through an LLM may or may not taint it; it may create sentences which on first glance are similar, but may have different, imprecise meaning - specific wording may be crucial in some cases. So if the style is under question, the content is as well. And if you can write the technically correct text at first, why would you put it through another step?
AI tools make different types of mistakes than humans, and that's a problem. We've spent eons creating systems to mitigate and correct human mistakes, which we don't have for the more subtle types of mistakes AI tends to make.
Presumably the "subject matter expert" will review the output of the LLM, just like a reviewer. I think it's disingenuous to assume that just because someone used AI they didn't look at or reviewed the output.
But why would a serious person claim that they wrote this without AI when it's obvious they used it?!
Using any tool is fine, but someone bragging about not having used a tool they actually used should make you suspicious about the amount of care that went to their work.
That’s fine. Write it out yourself and then ask an AI how it could be improved with a diff. Now you’ve given it double human review (once in creation then again reviewing the diff) and single AI review.
That's one review with several steps and some AI assistance. Checking your work twice is not equivalent to it having it reviewed by two people, part of reviewing your work (or the work of others) is checking multiple times and taking advantage of whatever tools are at your disposal.
> Once you do have good storyboards. You can easily do start-to-end GenAI video generation (hopping from scene to scene) and bring them to life and build your own small visual animated universes.
I keep hearing advocates of AI video generation talking at length about how easy the tools are to use and how great the results are, but I've yet to see anyone produce something meaningful that's coherent, consistent, and doesn't look like total slop.
I watched the most popular and most recent videos of each channel to compare, and they were all awful:
> Bots in the Hall
* voices don't match the mouth movements
* mouth movements are poorly animated
* hand/body movements are "fuzzy" with weird artifacts
* characters stare in the wrong direction when talking
* characters never move
* no scenes over 3 seconds in length between cuts
> Neural Viz
* animations and backgrounds are dull
* mouth movements are uncanny
* "dead eyes" when showing any emotions
* text and icons are poorly rendered
> The Meat Dept video for Igorrr's ADHD
This one I can excuse a bit since it's a music video, and for the sake of "artistic interpretation", but:
* continuation issues between shots
* inconsistent visual style across shots
* no shots longer then 4 seconds between cuts
* rendered text is illegible/nonsensical
* movement artifacts
You're just one person, those people have their own audiences; so your own critique, is just your own critique. Just because you don't like it, doesn't mean that it is not resonating well with others. I can tell you from the research I am doing for several hours per day on ai-filmmaking that there are already a few handful of creators making a living from this; with communities behind them that keep growing, and their audiences that keep expanding (some already have 100k to 1m subscribers across different social media channels). Some of them are even striking brand deals.
Entire narrative driven AI stores that are driven by AI stories and AI characters in AI generated universes... they are here already, but I can only count those who do it well on two hands (last year, there where 1-2). This is going to accelerate, and if you think its "slop" now, it just takes a few iterations of artists who you personally resonate with to jump onto this, before you stop seeing it as slop. I am jumping on this, because I can see very clearly where this will all lead. You don't have to like it, but it will arrive regardless.
Almost every talented artist with a public presence that has spoken on AI art, has spoken against it's generation, the use of AI tools, and the harm it's causing to their communities. The few established artists who are proponents of AI art (Lioba Brueckner comes to mind) have a financially incentive to do so, since they sell tools or courses teaching others with less/no talent to do the same.
The tools aren't going anywhere. Fans were outraged at the look and artists raged against the transition from cel animation to digital. Almost nothing serious is produced via cel now and the art adjusted by making extremely complex and beautiful art that couldn't have been done on cels.
There's a real legal fight that needs to go on right now about these companies stealing style, voices, likeness, etc. But it's really beginning to feel like there's a generation of artists that are hampering their career by saying they are above it instead of using the tools to enhance their art to create things they otherwise couldn't.
I see kids in high school using the tools like how I used Photoshop when I was younger. I see unemployed/under employed designers lamenting what the tools have done.
Why didn't you mention financial incentives of many outspoken critics of AI? They feel like their entire livelyhood depends on AI failing. I'd say that's a pretty strong financial incentive.
The issue for them is that once the tools exists, adoption only moves in one direction. And it will enable a whole wave of new artists. I sympathize with them, but if I enjoy GenAI art creation and see it as my genuine creative outlet, why would I stop? What about thousands of others exploring this?
If at some point I also get very good at it; and the tech, models and tools mature, this will turn into a real avenue; who are they to tell us not to pursue it?
I don’t think that is the problem (as someone that has been described in that bracket), it’s the tooling and control that is missing. I believe that will be solved over time.
Before industrial demolition was common, old buildings would be town down and material repurposed for new constructions, build on top of existing foundations and rubble. Do this enough over the centuries and your city will slowly rise in height.
If anyone’s ever in Barcelona I recommend checking out the history museum, which is literally built on top of some Roman and medieval ruins. You can descend into the basement to see the excavated remains of the foundations of Roman buildings that had been levelled and built on top of.
I think the author missed an important factor: misaligned incentives.
Dating apps make money when users spend time (and money) on the platform. Users who find a partner tend to leave the platform, so dating companies are incentivized to prevent that from happening. Those companies then have more opportunities to up-sell those users on premium features, which they're more likely to purchase due to repeated failure and/or feelings of inadequacy.
This is often parroted, but the reasoning is flawed. The vast majority of the platform's growth will come from new users, who are entering the dating scene. If they fail to capture that audience (say, by having a reputation of not performing as advertised), then no amount of upsells or string-alongs of existing users will sustain them, as their user base will only ever decrease, and investors will see that and withdraw accordingly.
1) The platforms aren't growing that impressively. Most of their users have been on the platform for a while, were previous users, etc.
2) It doesn't matter how good the app is, you need a network effect. New users are going to go to where the potential dates are.
3) Marketing does wonders. An app can suck and have great marketing. It will get users over an app that actually works and doesn't have good marketing.
4) Lots of people on dating apps are looking for dates (hookups), not partners. If the apps can keep you getting dates, not partners, they can keep you on the app and happy.
"If the apps can keep you getting dates, not partners, they can keep you on the app and happy."
I know this sounds judgemental but I'm not convinced the people going on lots of dates are "Happy" even if they're being successful in dating and hookups.
Match's growth peaked a long time ago. The site is now trying to grow by "offering new products" and "cutting operational costs."
The relative newcomers - Bumble and Hinge - grew by trying to offer a better experience, especially for women, who are traditionally overwhelmed with unreciprocated interest on conventional apps. Both seem to have admitted defeat now and moved to the usual model.
In terms of revenue, the incentive to keep millions of users spending is far higher than the nominal gains from persuading friends of a successful couple to join up. Given that most users aren't successful, that network effect is tiny.
There's an opposing network effect of *keeping customers unmatched, because this provides gossip and entertainment among friends, which gives them a reason to continue using a service.
We know that string-alongs are a real thing on dating sites - especially, but not exclusively, for men.
There's also a small but not negligible subculture of (mostly) women who use dates for free meals and get a good return on their monthly subscription.
And a lot of sites - not just Tinder - overlap hook-up culture with people seeking marriage and kids. If anything the former is a more popular option now.
FWIW, Hinge has been owned by match for some years now. Bumble is still independent by their stock is down ~92% over 5 years. I think they will eventually be bought out by match.
Wrong claim:
"There's an opposing network effect of *keeping customers unmatched, because this provides gossip and entertainment among friends, which gives them a reason to continue using a service."
No, this is implying they are doing it with intention - which they dont even have to insist on! They can keep the users, because matching based on an app does not work for 99.99% real cases.
So if you treat them well, they will stay anyway, unless your product is shitty.
I don't think this counterargument holds. It's a hell of a lot easier to get a customer who already paid once to pay a second time than it is to get a customer to pay for the first time. Also, I think most people are well aware that by and large, dating apps have a very low success rate for the majority of their users. People use them anyway.
Since your marginal costs per customer are veeeery low, you can hammer thm with 50-60% discount, which ends up on most platforms at 10 - 20 bucks per month, and if you make 3 - 4 dinners, you get much more out of it than an evening in the cinema
They don't need that. What option do most young people have?
Most young men can't approach women, most young women can't handle being approached and we don't have shared spaces where people can get to know each other and pair off anymore. Young people think the apps are dumpster fires, they hate them, but the alternative is sadly worse.
> The vast majority of the platform's growth will come from new users...
Userbase expansion is new users less leaving users for a time period. So there are two factors, not just "new users."
In any case, Match Group apps are well into the phase of focusing on extracting the most money possible from their paying users as opposed to gaining new users.
After all, infinite users are useless to a company, even if it costs nothing to support them, if none of them pay.
- i have worked in the space for some years for two of the biggest platforms in my country
- dating sites track a lot KPI and discuss them and test them thoroughly
- the KPI "do-more-users-leave-our-platform-earlier-if-our-matching-algo-is-just-too-good" - I promise: In alle the years, this question WAS NEVER - NEVER!!!!!!! - raised, regardless wich Manager or which Exec. This metric isnt even debated.
And here comes why:
the most important thing to form a relationship are "technicals" which can NEVER be introduced into an app. There may be some advances in genomic matching, but no body deployed this so far and it wont happen unless Apple watch as a gene encoding module.
There are one night stands, there some marriges (we had a "winners board" in our office), but 99% of all cases when people met, its going to be "failure" (in a sense "no match")
Regardless how good your algo is - it doesnt matter when it comes to a reality check.
Therefore, Dating apps have absolutely no fear of you signing off because you fond someome - its very likely that you will come back soon, second: From operators perspective it would be a good thing if people would tell "i found my match on XYZ", but sine this does happen only super rarely, there are only few such stories.
So - NO: Dating sites do fear someone deleting the account.
(except: you are a startup and have to keep every profile to gain some size)
> - the KPI "do-more-users-leave-our-platform-earlier-if-our-matching-algo-is-just-too-good" - I promise: In alle the years, this question WAS NEVER - NEVER!!!!!!! - raised, regardless wich Manager or which Exec. This metric isnt even debated.
What labels do they use for training their algos though? What is their definition of a successful match, is it a date, a recurring date, or something closer to a long-term relationship?
If matches predominantly result in "failure" they might just not have enough "long-term success" labels to go by, and their proxy labels will be biased towards short-term successes.
Wrong approach;
at least until latest, NONE of the standard apps does apply any kind of "real AI" stuff, maybe this changed through the last 3 - 4 years.
All thi matching stuff like "match with X%" is just bullshit.
The only platform having a useful approach here was OKC years ago. (but even for their scoring you would not need any type of sophisticated tech)
- new user sign up
- new subscriptions
- enlengthened subscriptions
- mails send
- overall interactions
- testing of new marketing campaigns/channels (they do massive MVT, even fontsize&coloring)
just to make that clear: churn is tracked/written down for sure, but its not debated in a context of "is churn growing if more people are matching successful" - because in 99.9% of all cases people do not meet successful, so this metric is irrelevant
And yet, the premium features are overtly aligned against user success!
I believe you that these app devs think they are optimizing for user success, but that doesn't change the incentives that frame their work.
The greatest utility of a dating app should be that it provides a higher number of opportunities. This feature is explicitly broken by the most popular dating apps. Often, it is put behind a paywall, which has the same effect as being broken.
Limitting search/result is often used to tease users into the subscription. Eg. Tiner allows in free mode only a certain number of swipes.
Is it this what you mean?
This is usually depends on: Search + text for free but limited, or "pay for everything"
I do not see why putting some features behind a paywall is "against users interest" and how this limits/increases his/her chances?
You are claiming that dating sites should be free - this doesnt work usually (POF as exception) - and making users pay for those does not increase his/her chances: EVERYTHING that happens before you met someone will be crashed usually in the very first second you met (and smell!) someone.
So if you have a chance with another person, is something that is completely(!) out of control of the website operator.
EDIT: this is something website operators do know, they cant change it and this is something that they should put on their website - they are selling dreams and expectations, which wont become true in real.
> Eg. Tiner allows in free mode only a certain number of swipes. Is it this what you mean?
Yes. The direct result of this is that other users have an arbitrarily limited number of opportunities to be swiped by you. A few users circumventing this limit (by paying to get around the wall) does not significantly raise this number.
Instead, it makes things worse, because a small number of paying users will get an outsized opportunity to swipe your profile! That's only useful if you are categorically attracted to people who pay for Tinder. Otherwise it is counterproductive to everyone involved.
This is exactly why I always make it a point to discourage my friends from using dating apps.
A dating app that is effective at solving the problem it is ostensibly designed to solve will never make money as people will be matched quickly and will have no need to pay for the service. So clearly no profitable dating app is good at matching people.
I'm of the opinion that using a tool that is constantly setting you up for romantic failure and rejection in the name of keeping you on its platform is a really good way to wreck your mental health.
On the other hand, if it was so easy to find a match, then we wouldn't be trying to use dating apps. I think it is just generally hard to find a good match for many people these days.
The other side of that is that many people are simply terrible and really unsuitable for being in any sort of long term stable relationship. No dating app can solve that problem (unless maybe they incorporate mental health and life coaching services, which seems kind of sketchy as a combination). Whether the situation has gotten generally harder these days is impossible to say but I certainly don't envy those still in the dating pool.
IMHO they should be the best one and use that to draw people to FB. They don't need to keep people in the dating app since its not the primary business.
Bingo. This is the effect that keeps (a) incumbent platforms in place, (b) users on those platforms, (c) and potentially new platforms from coming online and offering a "superior" experience.
This is pretty obviously addressed in the article. The premise of the article is why don't users migrate to better platforms when the large ones are extracting as much money as possible (because the incentives are misaligned)
Anyone can decide a statement is "ideological" just by disagreeing with it. The statement "the government can and should provide services to its people if those services are important to its future and the free market is incapable of providing them" should not, and did not used to be, ideological. The fact that such a statement is far outside of the current Overton Window is the result of a decades-long propaganda campaign to destroy everyone's faith in government so it can be looted, and everyone who parrots "government bad" is (knowingly or not) playing a part in this propaganda campaign. I assume the issue you have with my comment is that I knew 95% of readers would immediately regurgitate "but government bad!", but of course I was right that that happened.
When the Overton Window shifts to the point where saying "people should be decent to one another" becomes a radical ideological statement, make sure you flag every comment that says that too. We can't have radical ideologues on HN, after all.
The question of what goods and services should/shouldn't be provided by government is about as ideological as any question there is, given that the size/role of government has been one of the most central matters of most major ideologies for at least the past couple of centuries.
It's fine to discuss ideology on HN, but it needs to be done so in a spirit of curiosity and exploration, not battle and belligerence.
We've had to ask you to respect the site and observe the guidelines a few times before in the past couple of years. The HN guidelines and HN's intended purpose are the issue here, and sermonizing about ideology seems a deflection.
We just need you to observe the guidelines, no matter what point you're trying to make (and we really don't much care what point you're trying to make). Indeed this is in your interests to do this, because your point will carry more weight if you make it whilst keeping your discussion style within the site's guidelines and norms.
I’m sure there are many people who don’t want their dating lives influenced by whatever rulers are in charge of their governments. Democrats won’t like a MAGA dating service, and vice versa.
> This is exactly why a dating app should be developed and provided by the government. Side note: If this gives you the heebie jeebies, you are part of the problem.
Bravo, I haven't laughed this much in a while. God-tier satire.
> Did you know Iran never had any socialists or students in their ayatollah-dictatorship revolution?
"However, as ideological tensions persisted between Pahlavi and Khomeini, anti-government demonstrations began in October 1977, developing into a campaign of civil resistance that included communism, socialism, and Islamism."
> you will still find these facts if you dig into the detail pages on wikipedia
Yes, that's called "research"...
> the current leader of the UN, in his youth as a violent communist, used violence against LGBT protestors, or that he beat a few of his own students into the hospital
This sounds like a conspiracy theory, and I couldn't find any reference to those events anywhere online. Even the incredibly biased socialist/communist prolewiki doesn't name him as a communist: https://en.prolewiki.org/wiki/Ant%C3%B3nio_Guterres
> Did you know that North Korea is not communist? It is a "a highly centralized, one-party totalitarian dictatorship"
That's correct, the state ideology of north korea is "juche", which has it's roots in marxism/communism but splintered decades ago, with a focus on nationalism, historical revisionism and reverence for the leader(s). (https://en.wikipedia.org/wiki/Juche)
> Same for "socialist dictatorship", where you will only find explanations of Marxist theory, not the many disasters, some of which mentioned above.
The zillions of LOCs generated by AI tools are just job security for experienced developers, assuming enough are still around by the time companies realize the amount of debt they've buried themselves under.
I suspect that this depends on what the code does and where it lives. E.g. some unmaintainable web frontends can be simply discarded and get redesigned and rewritten rather than fixed.
https://far.computer/how-to/
reply