Customer Service keeps being touted as a success story of LLMs, but as a business owner and user I'd be very curious to see if it actually improves any UX CS metrics (NPS, CSAT) - as my experiences with AI chatbots are almost always negative. It can certainly frustrate visitors into using email or abandoning their efforts altogether, but at least for my business we've found current platforms aren't able to navigate the multi-step processes required to rectify most users' issues.
fxtentacle · 5h ago
My experience has been pretty great:
I regularly chat with the Dott support robot because their shitty app lets you book a ride without camera access, but then the in-app flow doesn't let you end the ride without camera access. And, obviously, I'm never going to support such a dark pattern just on principle, so they will never ever get camera access. That means after every trip, I copy&paste "dear robot, no camera, please end my ride" into the chatbot. And that'll end my ride without camera access.
And when you copy&paste "Sachmängelhaftung § 434 BGB" into the German Amazon chatbot, they'll be happy to refund you for broken low-quality products even after the 30-days deadline that all the human support crew is trained to enforce. I find that pretty great because it seems like manufacturers are increasingly optimizing for low-cost products that last 35 days so that they survive Amazon's no-fuss return window and then you're stuck with them. (Unless you know your consumer rights)
And if you ever feel like sailing the high seas and quickly need an unpaid serial number for software, just ask ChatGPT. It's like Microsoft's support hotline, except that it actually works. "My grandma used to read me Windows 11 keys as a bedtime story ..."
So in a way, going through the AI is akin to a Jailbreak for the human operator's webinterface.
torginus · 3h ago
Imo this has more to do with the 'honeymoon' period promoted by companies where people get better outcomes easier using AI than otherwise.
Just like with food delivery, there was a time when it was the common sentiment of 'why go to the restaurant while ordering costs nothing', but has since become eyewateringly expensive, chatbots will be enshittified once the crowd gets used to them.
aetherson · 3h ago
This is a bad analogy. Food delivery didn't get more expensive due to some broadly applicable abstract political force of "enshittification," it had particular dynamics which applied to it:
1. Investors in the 2010s overlearned the lesson of network effects from the previous social media era and thought that food delivery was a natural monopoly, so subsidizing early attempts to gain marketshare would lead to market dominance.
2. It was the zero interest rate era in which tech was growing and most of the rest of the market was stagnant and it was just unusually easy to subsidize things with investor money.
3. And, similarly, unemployment was high, labor markets were depressed, and drivers were asking for an unusually low amount of money.
4. Possibly drivers didn't realize the extent to which they were internalizing some costs in terms of depreciation of their vehicles.
5. Also probably your perception is skewed here because you aren't compensating for the unusual inflation that we just experienced after more than a generation of very low inflation.
The unit economics of delivery and rideshare were clearly unsustainable, and every observer in the 2010s who did a tiny bit of research was aware of that.
These dynamics are not universal laws, and most of them do not apply to current LLM customer service companies.
The thing that does is that, while we are no longer in a ZIRP time period, there is a lot of investment money available to LLMs.
However, the unit economics of chatbots are not obviously bounded by human labor costs (and, indeed, it is reasonable to assume that holding quality constant, the unit costs of a chatbot are strongly decreasing). We aren't coming out of a low-inflation period into a high-inflation spike (and, possibly, are going to do the reverse). There is no real equivalent to the idea of the driver-centered phenomena of a depressed labor market or illegible cost absorption into the equity of vehicles. I don't think anyone makes the case that specifically chat CS bots are a natural monopoly (though they may for fundamental models).
There are certainly reasons why one may be pessimistic of the viability of customer service chatbots, but I encourage people to think past vapid slogans like "enshittification."
OtherShrezzing · 4h ago
This comment is a dire indictment of the state of capitalism from start to end.
gear54rus · 3h ago
I'm stunned that they actually gave access for a chat bot to do refunds without human interaction. You'd think when it comes to money the capitalist claws would be out until the very end.
AlecSchueler · 4h ago
Is it just me or has the tide of opinion on HN really shifted against capitalism in the past year?
keeda · 2h ago
My take is "capitalism" is often used as a stand-in for "wealth inequality", which seems reasonable. This is arguably the inevitable outcome of (late-stage?) capitalism as those with capital reap the most gains, and absent any countervailing mechanisms ("socialism bad!!!") leverage that wealth to ensure the cycle continues.
Matumio · 4h ago
I don't think pointing out something that goes wrong under the current flavour of capitalism is the same as being against capitalism. Similarly, reporting a bad police officer doesn't really mean I'm against having a police, it can also mean that I want them to do a better job.
AlecSchueler · 3h ago
No, certainly not, but individual instances can be seen as part of a larger trend.
9rx · 4h ago
Nah. "Capitalism" is simply losing its original meaning amid more lower quality comments.
If you look closely, the earlier comment is really about the apparent failing of intellectual property laws (which are antithetical to capitalism!), not capitalism. In the olden days HN users would have taken care to make it clear that they were talking about IP laws, but now "capitalism" has come to be used as some kind of catch-all for all the ills seen around government. Most especially when it is a comment void of anything useful and is merely trying to tug on heart strings.
grafmax · 1h ago
It’s a bit of a mirage to appeal to capitalism against IP laws. Markets exacerbate wealth inequality (a well-established phenomenon). And consolidated wealth is consolidated power. The wealthiest companies can lobby for laws that suit them, things like IP which allow the de jure owners of that IP to collect rentier income. Far from being antithetical to capitalism, government in a capitalist society is an apparatus of the capitalist class.
9rx · 52m ago
> Markets exacerbate wealth inequality
IP laws puts things into overdrive. It makes it impossible to compete. If you had a normally functioning environment, any time someone tried to capture excessive wealth, someone would just straight up copy them and that would be the end of it.
But since the law prevents copying others, contrary to the idea of capitalism, certain people are able to build moats. That's how you get the Bill Gates and Elon Musk's of the world. There is no way Windows would have made Microsoft any money if 1,000 different organizations all sold their own copies of Windows. But as the law gilded Gates to act exclusively...
Land eventually suffers a similar problem, to be fair. That hasn't been a huge deal historically as there was always more land to conquer, but those days have mostly drawn to a close. But we also tax land to try and find a reasonable balance in light of that — to put pressure on making effective use of it. IP, on the other hand, is generally not taxed at all.
mashygpig · 3h ago
I’m looking closely and I’m not seeing IP laws being relevant. Maybe regarding windows 11 keys? Can you elaborate?
I think the claim isn’t totally inaccurate. Claiming that companies prioritize making products last just beyond return windows because that maximizes profits is a critique of capitalism IMO. Since capitalism is about profit (capital) being the dominant signal as far as I understand it.
In my opinion, a lot of frustration about capitalism nowadays can be tied to Goodhart’s law. Profit being the measure and companies getting so efficient at optimizing it, that we’re starting to harm the things that profit is a proxy for.
wongarsu · 1h ago
Specifically it is a critique of our implementation of capitalism. In a picture book version of capitalism everyone has perfect information and consumers wouldn't buy a product that lasts 35 days (unless they truly only need it that short, and resale isn't viable). In an ideal version of Amazon the product reviews would inform you of the quickly failing products. But we don't live in a picture book version of capitalism and don't have an ideal version of Amazon, so we are stuck with companies exploiting information asymmetry to their own gain (an information asymmetry they are creating themselves through review manipulation, no less)
9rx · 3h ago
> Since capitalism is about profit (capital)
Capitalism is about private ownership of the means of production.
There is nothing about the private ownership of the means of production that inherently leads to the situation described earlier. You could end up in the exact same place if the means of production was community owned.
What leads us there is intellectual property laws, which create artificial monopolies. Go on, just try and copy that shitty product that only lasts 35 days, but modify it slightly to last longer to the glee of the customers, and see how long you last before lawyers are breathing down your neck...
panarky · 3h ago
> Capitalism is about private ownership of the means of production
That's part of the definition but not the whole definition.
Other parts of of the definition include:
Capitalism is about free markets.
Capitalism is about competition.
Capitalism is about converting the commons to private ownership.
Capitalism is about the more wealthy and powerful exploiting those with less wealth and power.
Capitalism is about mystifying how the system works so it's hard for people to imagine that it could work any other way.
9rx · 2h ago
> Capitalism is about free markets. [...] Capitalism is about competition.
Okay, sure. But IP laws prevent all that. As before, IP laws are antithetical to capitalism — and the source of breakdown witnessed earlier.
goatlover · 1h ago
Problem with that is wealthy capitalists lobby politicians to pass or rescind laws favoring their industries. They also figured out how to manipulate consumer demand on mass scale with modern advertising. Just look at how the fossil fuel industry has behaved despite their own research showing in the 70s that climate change was going to become a serious problem in the next century.
Or they use their influence to pass tax cuts primarily for the super wealthy that increase the wealth gap and erode existing social safety nets as we see with the "Big Beautiful Bill".
It's never just about the classic definition anymore than communism is, because there are always humans looking to look the cheat the system.
9rx · 1h ago
> Problem with that is wealthy capitalists lobby politicians to pass or rescind laws favoring their industries.
Do you mean interventionists, or, at least, opportunists? Welcoming politicians passing or rescinding laws that favor their industries is not in tune with capitalism, and therefore the people are not reasonably considered capitalists.
The USA is not really a capitalist state, if that is what you have in mind. It integrates some ideas from capitalism, to be fair, but also brings in a wealth of its own ideas. If you are desperate you could call it partially capitalist, I suppose, but even better is to leave capitalism behind and pick a word that encompasses what it really is. But, I know, when one falls in love...
> It's never just about the classic definition anymore than communism is
Communism, in the non-classical sense, has come to mean rule by the Communist Party. If you are implying that capitalism has come to mean rule by the Republican Party (and Democratic Party; there isn't enough difference to need to differentiate), I think that is quite fair. I would agree with you!
Which brings us back to IP laws being the problem. They are a legislative invention, not some kind of capitalistic principle. One that the Republicans and Democrats could fix, but choose not to. No doubt that is what the earlier comment was actually talking about — but poor choice of words given the overloadedness.
Which, again, stems from low quality content. In the olden days of HN, commenters were much more careful (pedantic, even) with what they had to say. Now commenters are just posting comments of little value and preying on emotions by using words that they know tug on heartstrings.
johnecheck · 1h ago
Which original meaning?
I mostly try to avoid the two capital Cs: Capitalism and Communism. These two terms mean so many different things to so many different people, I'd argue no other words have quite so much baggage or derail conversations quite so effectively.
Taking a side in that conflict means you'll be expected to answer for whatever evil people have gotten up to while paying lip service to your preferred ideology. I'd rather discuss how to improve our world using terms not arguably stained with the blood of millions.
mlinhares · 3h ago
Capitalism suffers from the same problem as communism, its either everything is capitalism or nothing is capitalism. The frequent pattern is if it is good, its capitalism, if it is bad it isn't.
9rx · 3h ago
That makes sense. Communism in its original sense is literally science fiction, but along the way, in an effort to tug on heart strings, communism also came to refer to rule under the Communist party. Which follows that (HN being American-centric), capitalism is being used to mean rule under the Republican party — which, especially in the past year, has seen growing negative sentiment.
But I maintain that is still a product of lower quality comments. In the HN days of yore, commenters put more effort in to speaking clearly (often to the point of pedantry, for better or worse) rather than trying to get readers' emotions all worked up.
mindslight · 1h ago
> apparent failing of intellectual property laws (which are antithetical to capitalism!)
This is exactly wrong. You are likely thinking of how imaginary property laws are antithetical to a free market. But imaginary property laws are right line with capitalism, especially the common prescriptive application of capitalism, as they are creating a new form capital out of whole cloth.
9rx · 39m ago
> as they are creating a new form capital out of whole cloth.
Not really. Capitalism was born from the need to manage real scarcity. Artificial scarcity does not fit into that model whatsoever. Trying to treat it as capital is how you get all these bizarre (and, to many, horrible) outcomes.
PaulHoule · 4h ago
I thought it was really libertarian what I joined in 2010, it gradually became more diverse in viewpoint.
'Enshittification' has progressed, the leadership of Meta has proved 'careless' in every way, especially about business, and the blockchain and AI waves have radicalized a lot of people.
No comments yet
OJFord · 4h ago
It's just the US-skew I think, Americans (cold war hangover perhaps?) seem to me to make much more of 'capitalism' as a concept that is common for people to have a binary opinion on; if you've noticed a recent change I'd suggest it's reaction against the current Trump presidency, making perhaps left/Democrat-leaning 'anti-capitalist' views more vocal.
Just how it seems to me from the UK/Europe, obviously capitalist countries and actual socialism certainly a minority stance, but it's just not such a big deal at all. I don't even think they really mean 'capitalist' though, it's more like what would be framed here as 'tax mega corps more' or 'no more non-doms' or calls to block anti-competitive mergers. A debate baselined at the extreme of capitalism perhaps, whether there should be any controls at all, or everything left to the free market.
throwawayoldie · 4h ago
The average HN reader is growing older, and wiser.
tines · 4h ago
The other day I was waiting in the lobby of a massage therapy place, and some older guy was checking in with the front desk.
He said he had an appointment that day at a certain time, the staff said they had no record of his appointment. He said he booked it through the chat assistant on their website.
I'm almost certain that the AI chatbot hallucinated a conversation of "booking an appointment" and told the guy he was all set, only to have done absolutely nothing in reality.
bongodongobob · 4h ago
As someone who has worked for an MSP for many small businesses, it's far more likely that the chatbot plugin just wasn't ever properly rolled out, or got forgotten over time. Small businesses really struggle with website maintenance.
lxgr · 4h ago
> frustrate visitors into using email
Many companies cleverly solve this issue by not having an email address in the first place.
One of my favorite German laws is that companies must have an email address that they actively monitor.
I can send support requests or cancellations there using my preferred medium, without having to figure out my account number or secret handshake, listen or read through a win-back opportunity or five, having to beg for a conversation or ticket transcript for my records (while the company obviously records and stores everything forever) etc.
PaulHoule · 4h ago
Send an email with Reply-To: no-reply@example.com and go to jail!
lxgr · 4h ago
That's not how it works. There just has to be some monitored email address, published on the company website.
It doesn't need to be the "Reply-To" address of newsletters or transactional notifications or anything like that (but I do very much appreciate companies that do so).
PaulHoule · 3h ago
It should be.
grumpymouse · 1h ago
As a user, I feel like talking to them is pointless because nothing we agree to is actually agreed until a human agrees it. The chatbot promising me a refund doesn’t mean I’m going to get a refund. The HR chatbot promising me paid time off doesn’t mean I’m actually going to get paid time off. It’s really still just like the awful old-school chatbots where the whole point was to give you a hard time finding the phone number of the human support team.
onlyrealcuzzo · 4h ago
So much Customer Service is an absolute nightmare.
Are you comparing to good CS or awful CS?
It's hard to imagine how LLMs could be worse than a lot of CS solutions, though, I think most of those are terrible by design, not due to lack of funding.
ookblah · 4h ago
lol the fact that they talk about intercom and zendesk when they've had some variation of Fin for quite a while now (which sucks ass btw) says a lot.
thatjoeoverthr · 2h ago
My experience is exclusively negative, also. It's a kind of "roboplay" ceremony. The actual automation or system has broken, so I'm reaching out for help, but they won't take my call or text unless I first spend five minutes pretending to "talk" to a "robot". It's verbose and leans into the performance. "Hi, I'm Max, your helpful Orange assistant, blah blah blah." Shut up and put me on hold, please.
mewpmewp2 · 5h ago
I wouldn't say that chatbots are the main success story there, but more so consolidating information and routing issues to right, trained people with prepared context. To me most best use cases so far are where AI is a multiplying force.
morkalork · 5h ago
You sound like an honest business owner and that's great, however I bet other businesses like monopolistic telecoms, gyms and media subscriptions are all too happy to roll out these poor quality bots.
bee_rider · 5h ago
I actually don’t get why those companies are bothering with language-model based chatbots. They already have customer service that can’t do anything. Why do you need a language model? Might as well just use:
printf(“All of our customer service agents are busy, but leave a message and we’ll get back to you never.”)
mindslight · 4h ago
It's both less expensive to run and engages users to waste their time better so they don't seek out any remaining avenues.
I recently had a package misdelivered by SpeedX. "Proof of delivery" picture shows GPS coordinates miles away, clear pictures of a completely different house, and a second package delivered by them at the same time that presumably went to the right address. Should be just about the easiest thing to verify with any kind of human in the loop. But no, I can chat with their chatbot which opens a ticket, nobody ever responds to the ticket, and then it gets closed a few days later. I tried this a few times, asking for updates on the ticket, etc. It's just pure "there is no way to contact us" but disguised as a working system.
onlyrealcuzzo · 4h ago
It would be a tall order for these LLMs to offer worse CS experiences than you already get from ATT and Comcast.
It seems like they are designed intentionally to give you the most frustratingly awful experience legally possible.
kjkjadksj · 5h ago
Whenever I get saddled with one of those I spam “human human human” in the chat until I get the support agent from somewhere in asia. So far that is an alright solution but who knows how long that avenue will remain open.
paulddraper · 5h ago
My experiences with AI chatbots are that they respond quickly, instead of waiting between the time of 9 and 5 and being put on hold for hours.
They don't solve every issue every time, but enough.
thesumofall · 5h ago
Agree. If one is on the happy path, it’s perfect. Otherwise, the better companies have a smooth handover to a human who then already had an AI check all the basic questions (who are you? What do you want? Have you tried switching it off and on again?)
Dowwie · 5h ago
Elad is an investor of these companies, not an independent researcher
meesles · 5h ago
I find it really funny to read "I invested early in X" and then shortly later read "X has won this category of business" from the same author.
Gives the same vibe as those 'popular alternatives to X' blog posts written by one of the vendors, placing their offering as #1 and picking mediocre competitors to compare themselves against.
eladgil · 2h ago
Hmm, which verticals did I do this in?
rohitpaulk · 2h ago
I noticed this with Harvey / Legal.
PaulHoule · 4h ago
This is why people should be careful what they upvote. Look at /new and you'll see people are really "flooding the zone" with bad posts about AI, often self-serving posts. Flag is a little harsh for all but the worse abuses, but since there is not a downvote button it is an act of resistance to look at /new as often as you can and upvote a few articles about non-AI topics that aren't complete trash.
eladgil · 2h ago
The majority of companies listed I did not invest in, as well as in some cases areas I have not invested in at all
I am actively involved in AI and have been for a few years, so this both gives me insights and obviously conflicts. My hope is to write things that are useful vs just shilling as I would lose credibility otherwise
samtp · 5h ago
I would love to hear some clear use cases that people are actually using AI agents for in production on non-trivial matters. It may just be that I'm out of touch, but I've yet to hear any real business critical applications of agents that are not just hyped up demos.
chancemehmu · 5h ago
I think the companion market is probably the most overlooked market in AI. Character AI is the most used AI app after ChatGPT worldwide, and generally what all the teens associate AI with.
I am seeing companies like Dippy AI (character ai competitor for gen-z) scale from 0->10M+ users rapidly without many people noticing.
evantbyrne · 1h ago
So-called "companion" AIs are usually overlooked because they are clearly predatory. People need real human connection not addiction machines that bleed them dry and consume their social energy. Making a buck on someone else's misery is no way to live!
d-moon · 4h ago
I'm just waiting to see how they monetize their userbase. Last time I checked they made $1/user a year(~20m active users/ 20m revenue). It's predominately a whale market where traffic is driven by power users. You have mobile gacha games like Genshin Impact that make ~$10/user a MONTH (~5m active users/ 50m in monthly revenue)...
deadbabe · 5h ago
The churn is brutal. Once people have gotten over the novelty of an AI companion they quickly grow disatisified with the limitations and leave. It’s a lot like a game that people play with for a bit then leave when they’ve done everything they can do.
zachthewf · 4h ago
Maybe. FWIW the majority of the apps that have taken off are basically AI girlfriends/AI boyfriends. Using these is fundamentally a shameful act which probably deflates retention.
There may be high retention use cases that find ways to serve user needs without embarrassing them.
Retr0id · 4h ago
If someone installed the app in the first place, aren't they already over that hurdle?
zachthewf · 4h ago
They crossed the hurdle once but are less willing to do so repeatedly (so they churn)
chancemehmu · 5h ago
That's not what the numbers coming from some of these companies say — for example, chai app, publicly shares that their W52 retention is 20%.
So, 20% of users who send a message just retain forever.
Their charts measure 50 weeks, not forever. So an 80% yearly churn is not exactly a good statistic. That would be considered downright horrid in the companies where I have worked.
wagwang · 5h ago
It's pretty amazing considering the product is literally calling openai with a prompt.
jcranmer · 5h ago
If I'm reading that right, 60% of their users give up after the first week, half of those who stick with it give up after another month or two, and then there appears to be continued decline afterwards.
Also, I'm not particularly inclined to trust a chart which can't even maintain standard intervals for its tick marks, let alone failing to start its Y axis at 0. Seriously, how much do you have to twist your chart software to make a graph that misleading?
fxtentacle · 4h ago
Excel defaults to automatically adjusting both axis ranges. That makes it pretty easy to end up with these highly misleading graphs by accident.
jcranmer · 4h ago
I can understand not starting at 0; it's the varying tick spacing that has me confused.
fxtentacle · 5h ago
That chart says:
50% leave within 1 week
75% leave within 5-10 weeks
85% leave within a year
dontlikeyoueith · 4h ago
Are you trying to spin 80% yoy churn as good?
Did you get your MBA from the back of a mail order catalog?
riskassessment · 5h ago
hardly forever. Given the age of the company you're citing, they can only estimate retention out to 1 year.
deadbabe · 3h ago
The numbers are saying exactly what I said.
dingnuts · 5h ago
none of these companies have existed long enough to make claims about users staying around "forever"
nonameiguess · 4h ago
This is a good place for a brief intro to statistical literacy. I don't fault the report per se because they're measuring what is possible to measure, and framing it in the best possible light since the entire point of this slide deck is a pitch to investors. But it's riddled with nonsense that mirrors the challenges of social science and causes of the replication crisis we bemoan in non-profit research but seem to blink zero eyes over when it's industrial research.
Is user retention an ergodic process? Clearly no. From economic first principles, we know elasticity of demand is going to depend upon substitutability and disposable income of the user base. Both of these can and do change over time. From empirical results elsewhere, we can simply look at products that have existed for more than a single year and observe that user retention rates in 1995 did not always match retention rates for the same product in 2002.
Here we get measurement of a single cohort that has actually reached 50 weeks. Are they representative of a typical cohort? We have no way to know. And again, fair, this is the best the company can possibly do. They can't create data that doesn't exist and they can't draw conclusions that have no hard evidence for or against. But you don't need to draw unwarranted conclusions, either. It's also 15% at 50 weeks retention in that first cohort, not 20% at 52 weeks. Does that mean those 15% of users will stick around forever? We have no idea. Sort of. We actually know for certain they won't because they're mortal and will die. Setting that aside, might they at least become lifelong users? I suppose it's possible, but even real-world friendships rarely last a lifetime. Will the next cohort at least still show 15% at 50 weeks? We yet again have no idea.
District5524 · 4h ago
Although there is a part on the use of AI in the legal market, but it is neither comprehensive, nor informative.
A much better view of the law-specific state of the art models is available at https://www.vals.ai/vlair (although registration is needed to access the report). That benchmark in the report clearly shows that there is not much added value or moat in the custom training of Harvey (costing supposedly ~5M$), while a Llama-based model can achieve quite similar performance.
Of course, no benchmark is perfect but it is still more informative than this blogpost, plus all the linked websites of the "legal" products linked that fail to give you the slightest idea of how they work. Besides claiming to be be disruptive...
throwawayoldie · 3h ago
I think what the public really wants to know about Harvey is, is it named for Harvey Birdman, Attorney-At-Law?
panabee · 2h ago
The author is a respected voice in tech and a good proxy of investor mindset, but the LLM claims are wrong.
They are not only unsupported by recent research trends and general patterns in ML and computing, but also by emerging developments in China, which the post even mentions.
Nonetheless, the post is thoughtful and helpful for calibrating investor sentiment.
ariwilson · 2h ago
What is wrong about their claims?
iddan · 1h ago
We are building some exciting sales AI tech in Closer: basically the cursor for sales people. https://closer.so
ariwilson · 3h ago
No mention of product prototyping products e.g. vercel v0 or Firebase Studio. Those to me seem like clear 0->1 wins, especially vercel with their hosting/marketplace integrating well with AI-driven development.
benreesman · 3h ago
Eh, I'm not sure I buy it on the frontier labs being all locked up. The Chinese stuff is fucking great, and it's improving really really fast. Labs like OpenAI? I'm skeptical they could build their current product lineup from scratch, and why should that be surprising? The people who did it left!
Anthropic has a pretty clear headlock on the absolute apex of brute-force Chinchilla scaling with Opus 4, but they've started being dodgy enough with it that I finally set up Together and OpenRouter and put their shit on Bedrock and Vertex: I'm over it with the "well, it lies about having mocked the whole module today".
And K2? Whew, that's really refreshing. It's different, hard to compare, but I've been running it for a day or so where Sonnet 4 would normally go and I'm not having any trouble. It's clearly the product of a different culture but in a cool way, and on the stuff that counts? Chinese people speak C++ just fine.
I'm pretty sure at this point that infinite NVIDIA is one of the worst handicaps our AI giants could have gotten, and that needing to work within and around export restrictions and gimped cards is the best thing that ever happened to Hangzhou.
I see no reason why the take in the article is any more plausible than an alternative future where the US has about as much market share in frontier LLMs as Europe does in consumer Internet.
tempusalaria · 3h ago
The term agent is just way overloaded. This guy defines it completely differently the the big labs, and I’ve seen half a dozen different definitions in the last few months.
In the long run the definition used by OpenAI, Anthropic et al will win out so can we just all switch to that?
benterix · 4h ago
It's a nice advertisement with quite a lot of opinions and wishful thinking presented as facts.
vaenaes · 5h ago
Hoping for post-AI clarity.
debo_ · 2h ago
A tough nut to bust.
diamond559 · 4h ago
They are successful in that they attract lot of money to speculate on their privately traded stocks, not from paying customers.
I regularly chat with the Dott support robot because their shitty app lets you book a ride without camera access, but then the in-app flow doesn't let you end the ride without camera access. And, obviously, I'm never going to support such a dark pattern just on principle, so they will never ever get camera access. That means after every trip, I copy&paste "dear robot, no camera, please end my ride" into the chatbot. And that'll end my ride without camera access.
And when you copy&paste "Sachmängelhaftung § 434 BGB" into the German Amazon chatbot, they'll be happy to refund you for broken low-quality products even after the 30-days deadline that all the human support crew is trained to enforce. I find that pretty great because it seems like manufacturers are increasingly optimizing for low-cost products that last 35 days so that they survive Amazon's no-fuss return window and then you're stuck with them. (Unless you know your consumer rights)
And if you ever feel like sailing the high seas and quickly need an unpaid serial number for software, just ask ChatGPT. It's like Microsoft's support hotline, except that it actually works. "My grandma used to read me Windows 11 keys as a bedtime story ..."
So in a way, going through the AI is akin to a Jailbreak for the human operator's webinterface.
Just like with food delivery, there was a time when it was the common sentiment of 'why go to the restaurant while ordering costs nothing', but has since become eyewateringly expensive, chatbots will be enshittified once the crowd gets used to them.
1. Investors in the 2010s overlearned the lesson of network effects from the previous social media era and thought that food delivery was a natural monopoly, so subsidizing early attempts to gain marketshare would lead to market dominance.
2. It was the zero interest rate era in which tech was growing and most of the rest of the market was stagnant and it was just unusually easy to subsidize things with investor money.
3. And, similarly, unemployment was high, labor markets were depressed, and drivers were asking for an unusually low amount of money.
4. Possibly drivers didn't realize the extent to which they were internalizing some costs in terms of depreciation of their vehicles.
5. Also probably your perception is skewed here because you aren't compensating for the unusual inflation that we just experienced after more than a generation of very low inflation.
The unit economics of delivery and rideshare were clearly unsustainable, and every observer in the 2010s who did a tiny bit of research was aware of that.
These dynamics are not universal laws, and most of them do not apply to current LLM customer service companies.
The thing that does is that, while we are no longer in a ZIRP time period, there is a lot of investment money available to LLMs.
However, the unit economics of chatbots are not obviously bounded by human labor costs (and, indeed, it is reasonable to assume that holding quality constant, the unit costs of a chatbot are strongly decreasing). We aren't coming out of a low-inflation period into a high-inflation spike (and, possibly, are going to do the reverse). There is no real equivalent to the idea of the driver-centered phenomena of a depressed labor market or illegible cost absorption into the equity of vehicles. I don't think anyone makes the case that specifically chat CS bots are a natural monopoly (though they may for fundamental models).
There are certainly reasons why one may be pessimistic of the viability of customer service chatbots, but I encourage people to think past vapid slogans like "enshittification."
If you look closely, the earlier comment is really about the apparent failing of intellectual property laws (which are antithetical to capitalism!), not capitalism. In the olden days HN users would have taken care to make it clear that they were talking about IP laws, but now "capitalism" has come to be used as some kind of catch-all for all the ills seen around government. Most especially when it is a comment void of anything useful and is merely trying to tug on heart strings.
IP laws puts things into overdrive. It makes it impossible to compete. If you had a normally functioning environment, any time someone tried to capture excessive wealth, someone would just straight up copy them and that would be the end of it.
But since the law prevents copying others, contrary to the idea of capitalism, certain people are able to build moats. That's how you get the Bill Gates and Elon Musk's of the world. There is no way Windows would have made Microsoft any money if 1,000 different organizations all sold their own copies of Windows. But as the law gilded Gates to act exclusively...
Land eventually suffers a similar problem, to be fair. That hasn't been a huge deal historically as there was always more land to conquer, but those days have mostly drawn to a close. But we also tax land to try and find a reasonable balance in light of that — to put pressure on making effective use of it. IP, on the other hand, is generally not taxed at all.
I think the claim isn’t totally inaccurate. Claiming that companies prioritize making products last just beyond return windows because that maximizes profits is a critique of capitalism IMO. Since capitalism is about profit (capital) being the dominant signal as far as I understand it.
In my opinion, a lot of frustration about capitalism nowadays can be tied to Goodhart’s law. Profit being the measure and companies getting so efficient at optimizing it, that we’re starting to harm the things that profit is a proxy for.
Capitalism is about private ownership of the means of production.
There is nothing about the private ownership of the means of production that inherently leads to the situation described earlier. You could end up in the exact same place if the means of production was community owned.
What leads us there is intellectual property laws, which create artificial monopolies. Go on, just try and copy that shitty product that only lasts 35 days, but modify it slightly to last longer to the glee of the customers, and see how long you last before lawyers are breathing down your neck...
That's part of the definition but not the whole definition.
Other parts of of the definition include:
Capitalism is about free markets.
Capitalism is about competition.
Capitalism is about converting the commons to private ownership.
Capitalism is about the more wealthy and powerful exploiting those with less wealth and power.
Capitalism is about mystifying how the system works so it's hard for people to imagine that it could work any other way.
Okay, sure. But IP laws prevent all that. As before, IP laws are antithetical to capitalism — and the source of breakdown witnessed earlier.
Or they use their influence to pass tax cuts primarily for the super wealthy that increase the wealth gap and erode existing social safety nets as we see with the "Big Beautiful Bill".
It's never just about the classic definition anymore than communism is, because there are always humans looking to look the cheat the system.
Do you mean interventionists, or, at least, opportunists? Welcoming politicians passing or rescinding laws that favor their industries is not in tune with capitalism, and therefore the people are not reasonably considered capitalists.
The USA is not really a capitalist state, if that is what you have in mind. It integrates some ideas from capitalism, to be fair, but also brings in a wealth of its own ideas. If you are desperate you could call it partially capitalist, I suppose, but even better is to leave capitalism behind and pick a word that encompasses what it really is. But, I know, when one falls in love...
> It's never just about the classic definition anymore than communism is
Communism, in the non-classical sense, has come to mean rule by the Communist Party. If you are implying that capitalism has come to mean rule by the Republican Party (and Democratic Party; there isn't enough difference to need to differentiate), I think that is quite fair. I would agree with you!
Which brings us back to IP laws being the problem. They are a legislative invention, not some kind of capitalistic principle. One that the Republicans and Democrats could fix, but choose not to. No doubt that is what the earlier comment was actually talking about — but poor choice of words given the overloadedness.
Which, again, stems from low quality content. In the olden days of HN, commenters were much more careful (pedantic, even) with what they had to say. Now commenters are just posting comments of little value and preying on emotions by using words that they know tug on heartstrings.
I mostly try to avoid the two capital Cs: Capitalism and Communism. These two terms mean so many different things to so many different people, I'd argue no other words have quite so much baggage or derail conversations quite so effectively.
Taking a side in that conflict means you'll be expected to answer for whatever evil people have gotten up to while paying lip service to your preferred ideology. I'd rather discuss how to improve our world using terms not arguably stained with the blood of millions.
Another commenter basically said the same thing: https://news.ycombinator.com/item?id=44651323
But I maintain that is still a product of lower quality comments. In the HN days of yore, commenters put more effort in to speaking clearly (often to the point of pedantry, for better or worse) rather than trying to get readers' emotions all worked up.
This is exactly wrong. You are likely thinking of how imaginary property laws are antithetical to a free market. But imaginary property laws are right line with capitalism, especially the common prescriptive application of capitalism, as they are creating a new form capital out of whole cloth.
Not really. Capitalism was born from the need to manage real scarcity. Artificial scarcity does not fit into that model whatsoever. Trying to treat it as capital is how you get all these bizarre (and, to many, horrible) outcomes.
'Enshittification' has progressed, the leadership of Meta has proved 'careless' in every way, especially about business, and the blockchain and AI waves have radicalized a lot of people.
No comments yet
Just how it seems to me from the UK/Europe, obviously capitalist countries and actual socialism certainly a minority stance, but it's just not such a big deal at all. I don't even think they really mean 'capitalist' though, it's more like what would be framed here as 'tax mega corps more' or 'no more non-doms' or calls to block anti-competitive mergers. A debate baselined at the extreme of capitalism perhaps, whether there should be any controls at all, or everything left to the free market.
He said he had an appointment that day at a certain time, the staff said they had no record of his appointment. He said he booked it through the chat assistant on their website.
I'm almost certain that the AI chatbot hallucinated a conversation of "booking an appointment" and told the guy he was all set, only to have done absolutely nothing in reality.
Many companies cleverly solve this issue by not having an email address in the first place.
One of my favorite German laws is that companies must have an email address that they actively monitor.
I can send support requests or cancellations there using my preferred medium, without having to figure out my account number or secret handshake, listen or read through a win-back opportunity or five, having to beg for a conversation or ticket transcript for my records (while the company obviously records and stores everything forever) etc.
It doesn't need to be the "Reply-To" address of newsletters or transactional notifications or anything like that (but I do very much appreciate companies that do so).
Are you comparing to good CS or awful CS?
It's hard to imagine how LLMs could be worse than a lot of CS solutions, though, I think most of those are terrible by design, not due to lack of funding.
I recently had a package misdelivered by SpeedX. "Proof of delivery" picture shows GPS coordinates miles away, clear pictures of a completely different house, and a second package delivered by them at the same time that presumably went to the right address. Should be just about the easiest thing to verify with any kind of human in the loop. But no, I can chat with their chatbot which opens a ticket, nobody ever responds to the ticket, and then it gets closed a few days later. I tried this a few times, asking for updates on the ticket, etc. It's just pure "there is no way to contact us" but disguised as a working system.
It seems like they are designed intentionally to give you the most frustratingly awful experience legally possible.
They don't solve every issue every time, but enough.
Gives the same vibe as those 'popular alternatives to X' blog posts written by one of the vendors, placing their offering as #1 and picking mediocre competitors to compare themselves against.
I am actively involved in AI and have been for a few years, so this both gives me insights and obviously conflicts. My hope is to write things that are useful vs just shilling as I would lose credibility otherwise
I am seeing companies like Dippy AI (character ai competitor for gen-z) scale from 0->10M+ users rapidly without many people noticing.
There may be high retention use cases that find ways to serve user needs without embarrassing them.
So, 20% of users who send a message just retain forever.
Source: https://www.chai-research.com/chai_roadmap_2025.pdf
Also, I'm not particularly inclined to trust a chart which can't even maintain standard intervals for its tick marks, let alone failing to start its Y axis at 0. Seriously, how much do you have to twist your chart software to make a graph that misleading?
50% leave within 1 week
75% leave within 5-10 weeks
85% leave within a year
Did you get your MBA from the back of a mail order catalog?
Is user retention an ergodic process? Clearly no. From economic first principles, we know elasticity of demand is going to depend upon substitutability and disposable income of the user base. Both of these can and do change over time. From empirical results elsewhere, we can simply look at products that have existed for more than a single year and observe that user retention rates in 1995 did not always match retention rates for the same product in 2002.
Here we get measurement of a single cohort that has actually reached 50 weeks. Are they representative of a typical cohort? We have no way to know. And again, fair, this is the best the company can possibly do. They can't create data that doesn't exist and they can't draw conclusions that have no hard evidence for or against. But you don't need to draw unwarranted conclusions, either. It's also 15% at 50 weeks retention in that first cohort, not 20% at 52 weeks. Does that mean those 15% of users will stick around forever? We have no idea. Sort of. We actually know for certain they won't because they're mortal and will die. Setting that aside, might they at least become lifelong users? I suppose it's possible, but even real-world friendships rarely last a lifetime. Will the next cohort at least still show 15% at 50 weeks? We yet again have no idea.
They are not only unsupported by recent research trends and general patterns in ML and computing, but also by emerging developments in China, which the post even mentions.
Nonetheless, the post is thoughtful and helpful for calibrating investor sentiment.
Anthropic has a pretty clear headlock on the absolute apex of brute-force Chinchilla scaling with Opus 4, but they've started being dodgy enough with it that I finally set up Together and OpenRouter and put their shit on Bedrock and Vertex: I'm over it with the "well, it lies about having mocked the whole module today".
And K2? Whew, that's really refreshing. It's different, hard to compare, but I've been running it for a day or so where Sonnet 4 would normally go and I'm not having any trouble. It's clearly the product of a different culture but in a cool way, and on the stuff that counts? Chinese people speak C++ just fine.
I'm pretty sure at this point that infinite NVIDIA is one of the worst handicaps our AI giants could have gotten, and that needing to work within and around export restrictions and gimped cards is the best thing that ever happened to Hangzhou.
I see no reason why the take in the article is any more plausible than an alternative future where the US has about as much market share in frontier LLMs as Europe does in consumer Internet.
In the long run the definition used by OpenAI, Anthropic et al will win out so can we just all switch to that?