Stop selling “unlimited”, when you mean “until we change our minds”
328 comments
·July 29, 2025epistasis
energy123
Google and OpenAI have done similar things with their LLM offerings.
Gemini Advanced offered 2.5 Pro with nearly unlimited rate limits, then nerfed it to 100/day.
OpenAI silently nerfed the maximum context window of reasoning models in their Pro plan.
Accompanying the nerf is usually a psy op, like nerfing to 50/day then increasing it to 100/day so the anchoring effect reduces the grievance.
It's a smart ploy because as much as we like to say there's no moat, the user does face provider switching costs (time and effort), which serves as a mini-moat for status quo provider.
So providers have an incentive to rope people in with a loss leader, and then rug pull once they gained market share. Maybe 40% of the top 5% of Claude users are now too accustomed to their Claude-based workflows, and inertia will keep them as customers, but now they're using the more expensive API instead. Anthropic won.
Modern bait and switch, although done intelligently so no laws are broken.
epistasis
I had been loyal, but am not any longer, so the ploy definitely did not work with me. I guess I'll move on to Gemini now, until I get sick of it.
To the degree there is a moat, I do not think it will be effective at keeping people in. I had already been somewhat disillusioned with the AI hype, but now I am also disillusioned with the company who I thought was the best actor in the space. I am happy that there is unlikely to be a dominant single winner like there was for web search or for operating systems. That is, unless there's a significant technological jump, rather than the same gradual improvement that all the AI companies are making.
__MatrixMan__
Loyalty is for henchmen.
Bluestein
> I had already been somewhat disillusioned with the AI hype, but now I am also disillusioned with the company
Likewise: a faulty, unproven, hallucinating, error-prone service, however good, was a good value at approx 25 USD/month in an "absolutely all you can eat", wholesale regime ...
... now? Reputational risk aside, they force their users to appraise their offering in terms of actual value offered, in the market.-
fleebee
I don't even want to imagine how bad it will get if a legitimate moat does surface and users get entrenched ever deeper into the ecosystem of a single provider. A lot of the companies in this space have a track record of squeezing as much value out of their customers as they can get away with.
486sx33
[dead]
londons_explore
> the user does face provider switching costs (time and effort), which serves as a mini-moat for status quo provider.
When a provider gets memory working well, I expect them to use this to be a huge moat - ie. they won't let you migrate the memories, because rather than being human readable words they'll be unintelligible vectors.
I imagine they'll do the same via API so that the network has a memory of all previous requests for the same user.
ileonichwiesz
Is memory all that useful for using these LLMs? I’ve found that I mostly use them for discrete tasks - helping me learn a specific thing, build a specific project, debug a specific piece of code, and once it’s done I’d actually prefer it to forget that thing instead of keeping it around forever.
Hell, “just open a new chat and start over” is an important tool in the toolbox when using these models. I can’t imagine a more frustrating experience than opening a new chat to try something from scratch only for it to reply based on the previous prompt that I messed up.
andyferris
Unintelligible vectors might be hard to transfer from one of their older models to one of their newer models - so I think the human readable words will remain a bit of a narrow waist in this space for the immediate future at least.
ibaikov
I wonder if that's actually illegal, because it feels very close to false ads etc. It seems legal, but I think courts would side with customers.
As if google would say that yes, emails are $5/mo, but there's actually a limit on number of emails daily, and also number of characters in the email. It just feels so illegal to nerf a product that much.
Same with AI companies changing routing and making models dumber from time to time.
aspenmayer
If there is a material difference in the product that causes you to no longer feel that it's the same as it was when you subscribed, it could be considered a bait and switch? I think as soon as you notice that this is the case, you should probably stop paying them though, otherwise you might seem to accept this state of affairs. If you had a long term contract that didn't have some kind of language that tried to prevent this from happening in the first place, you could probably get out from under that contract by saying that the deal has essentially changed out from under you, but I think a lawyer might make that argument much better than me.
I'm not sure what harm you think you're suffering from, and what a proper remedy might be, if you think it's illegal. I don't know if I would go that far, as there are all kinds of words most terms of service use to somehow make it so that you have already acknowledged and agreed to whatever they decide to do. So a lawyer will probably be helpful there as well.
nojs
It’s not just switching costs though, what alternatives to Claude Code are as good right now?
danskeren
Claude definitely engage in some shady and illegal behavior. I wanted to see what the annual plan would cost as it was just displaying €170+VAT, and when I clicked the upgrade button to find out (I checked everywhere on the page) then I was automatically subscribed without any confirmation and without ever seeing the final price before the transaction was completed. Turns out that price was €206.50. Getting refunded was also a pain in the ass.
ponyous
€170 + 21.5% (Irish VAT rate) is €206.55. So not sure what you expected. One complain though in EU B2C companies must show prices with VAT included.
qilo
By clicking, he expected to see the full price with VAT included, as required by EU regulations, without doing the math himself.
diggan
> €170 + 21.5% (Irish VAT rate) is €206.55. So not sure what you expected.
Parent clearly stated they only saw "€170+VAT" and not €206.55, so of course they expected to see €206.55 before the purchase went through. Not sure what anyone else would expect?
OtherShrezzing
You might not have noticed in all the frustration, but you got overcharged. 170 + 20% VAT is 204, not 206.50.
Maybe they added a card fee in at the end, but if they didn’t make that abundantly clear, they’ve broken a law in most countries which use the Euro.
Etheryte
VAT is not 20% everywhere worldwide.
Insanity
Why 20% VAT? Different EU countries have different tax brackets, so OP might not have had 20%.
jrs235
Went to see if I could cancel my Claude pro plan just now. On billing I see adjust plan button. On that page I can upgrade but no where can I find downgrade/cancel. I checked the account page. I see delete account but "To delete your account, please cancel your Claude Pro subscription first."
Update: below the fold at the bottom of the Billing page is the cancel section and cancel button.
Update 2: just clicked cancel and was offered a promo of 20% off for three months...
Update 3: FYI, I logged in to my Claude account via computer (not iOS or Android).
93po
i tried this to see if i could get a discount and it just canceled it, lol. i wonder if its bc im a heavier user?
Waterluvian
My credit card web portal has a section for reversing transactions without having to call anyone. Maybe yours does too?
lukan
Reversing transactions is not cancelling the service I believe?
As long as you don't cancel, you do owe them money. But if they make cancelling intentionally hard, one would likely have a good case in court to still not pay, if one would want to go to court over this.
jonathantf2
My bank app has a bit where it detects common services and tells you how to cancel them in the app, like Amazon Prime
pnvdr
nice feature. so its like backcharge via app instead of calling credit card customer care. do you mind sharing the credit card company
Renevith
This is getting to be the norm rather than a unique feature. I can dispute a credit card transaction through the app for my Citi, X1, or Fidelity cards.
Waterluvian
Tangerine Mastercard Canada.
null
benreesman
I've now got it behind OpenRouter on Bedrock and Vertex and am rotating in K2 and Qwen more and more as I learn how to get the most of them because the silent degredation and work avoidance heuristics are not a variable I need in an already difficult expectation management regime to my stakeholders.
At the rate the Chinese are going it won't be long before I can shake the dust off my sandals of this bullshit for good.
avereveard
Qwen3 coder is very good and a tenth of clcaude price on openrouter. And has a :free version for light usage.
I still revert to gemini pro 2.5 here and there and claude for specific demanding tasks, but bulk token go trough open weight model at the moment.
93po
i use privacy dot com cards to prevent this issue, takes 2 seconds to turn off the card in my extension
null
conartist6
"I firmly believe that AI will not replace developers, but a developer using AI will replace a developer who does not."
Ugh, anyone who says that and really believes it can no longer see common sense through the hype goggles.
It's just stupid and completely 100% wrong, like saying all musicians will use autotune in the future because it makes the music better.
It's the same as betting that there will be no new inventions, no new art, no works of genius unless the creator is taking vitamin C pills.
It's one of the most un-serious claims I can imagine making. It automatically marks the speaker as a clown divorced from basic facts about human ability
atonse
I disagree. While there are developers that truly build new technology and invent novel solutions, the overwhelming majority of developers who are paid for their work daily do pretty mundane and boring software development. They are implementing business logic, building forms, building tables, etc.
And AI already excels at building those sorts of things faster and with cleaner code. I’ve never once seen a model generate code that’s as ugly and unreadable as a lot of the low quality code I’ve seen in my career (especially from Salesforce “devs” for example)
And even the ones that do the more creative problem solving can benefit from AI agents helping with research, documentation, data migration scripts, etc.
conartist6
I'm one working on novel tech, without AI. I've never been doing more valuable work or been more in command of my craft.
Yet the blanket statement is that I will fail and be replaced, and in fact that people like me don't exist!
So heck yeah I'll come clap back on that.
jennyholzer
Continuing to develop my craft in the 4th straight year of numbskulls claiming that Chat GPT is the end of history has made me extraordinarily confident in my ability to secure lucrative developer work for the rest of my life.
infecto
Your interpretation of the original quote is so far off that I would question your interpretation of the world. Sure novel stuff is being built but the vast majority of code at all sizes of companies has been written before in some iteration. Even for the novel work being done, it’s being surrounded by layers of code that has probably being written before. Are engineers going anywhere? No. But I also don’t think it’s far fetched to see a possible near term of competent engineers who use AI tools being more productive than the ones who don’t. I am not talking about copy and pasting but rather thoughtful use of tooling.
atonse
Right, I chose my words carefully when I said the "overwhelming majority" – and not "every single developer"
apwell23
> the overwhelming majority of developers who are paid for their work daily do pretty mundane and boring software developmet
So are musicians. We think of them as doing creative stuff but a vast majority is mundane.
staunton
Most musicians (i.e. non-famous ones) get most of their income from teaching students. I don't think such a model makes sense for developers
(though who knows, maybe at some time in the future there will be significant numbers of people programming as a hobby and wanting to be coached by a human...)
NiloCK
Do the people in this corner use compilers? Would they agree that programmers who don't use them* have been replaced by those that do?
*: I'm aware of cases like the recent ffmpg assembly usage that gave a big performance boost. When talking about industrial trend lines, I'm OK with admitting 0.001% exceptions.
(Apologies if it comes across as snarky or pat, but I honestly think the comparison is reasonable.)
nottorp
> Do the people in this corner use compilers? Would they agree that programmers who don't use them* have been replaced by those that do?
Are you aware compilers are deterministic most of the time?
If a compiler had a 10% chance of erasing your code instead of generating an executable you'd see more people still using assembly.
conartist6
Compilers are systems that tame complexity in the "grug-brain" sense. They're true extensions of our senses and the information they offer should be provably correct.
The basic nature of my job is to maintain the tallest tower of complexity I can without it falling over, so I need to take complexity and find ways to confine it to places where I have some way of knowing that it can't hurt me. LLMs just don't do that. A leaky abstraction is just a layer of indirection, while a true abstraction (like a properly implemented high-level language) is among the most valuable things in CS. Programming is theory-building!
bee_rider
Compilers were invented pretty early on in things… I wouldn’t be they shocked if the population of assembly programmers has remained constant.
Where would you put the peak? Fortran was invented in the 50’s. The total population of programmers was tiny back then…
tzumaoli
When Fortran came out, I don't think a lot of people yelled at the assembly programmers and told them "learn Fortran or be replaced".
jennyholzer
the comparison is preposterous.
diabllicseagull
this is kind of a funny example to me because of all the programming language and compiler discourse that's happening. analogies almost always miss the mark by hiding the nuances that need discussing, and this topic is no exception.
wat10000
There are technologies that become de facto requirements for work in a field. For software, compilers and version control both qualify.
But... what else? These things are rare. It’s not like there’s a new thing that comes along every few years and we all have to jump on or be left behind, and LLMs are the latest. There’s definitely a new thing that comes along every few years and people say we have to jump on or be left behind, but it almost never bears out. Many of those ended up being useful, but not essential.
I see no indication that LLMs or associated tooling are going to be like compilers and version control where you pretty much can’t find anyone making a living in the field without them. I can see them being like IDEs or debuggers or linters where they can be handy but plenty of people do fine without them.
tirumario
Full disclosure, I'm on the Kilo Code team, but I read your analogy, and I have to respectfully disagree. Musicians don't all use autotune, because autotune is a specalized technology used to elicit a specific result. BUT, MOST musicians use technology; either to record their work, or mix their tracks, or promote their work. You could definitely say "A musician who doesn't post online to a platform or save their work in certain audio formats at the studio are going to be replaced by musicians who do." Are there musicians who still release their work on vinyl or cassette tapes and prefer the sound of an stage with no microphones? Sure. But to dismiss the overarching influence of technology on the process would be ignoring where the progress is going. I'd argue that people who use Kilo Code don't just "autotune" their code, they're using it as a tool that augments their workflow and lets them build more, faster. That's valuable to an employer. Where the engineer is still vital is in their ability to know what to ask for, how to ask for it, and how to correct the tool when it's wrong, because it'll never been 100% right. It's just not hype, it's inevitable.
conartist6
I actually agree with you that LLM assistance is inevitable. The fact that we can have small local models is what convinces me that the tech won't go away.
Even if things are going the direction you say, though, Kilo is still just a fork of VSCode. Lipstick on a pig, perhaps. I would bet that I know the strengths and weaknesses of your architecture quite a lot better than anyone on the Kilo team because the price of admission for you is not questioning any of VSCode's decisions, while I consider all of them worthy of questioning and have done so at great length in the process of building something from scratch that your team bypassed.
LeafItAlone
I believe it. Maybe not replace 100%, but effectively replace it.
I believe that at some point, AI will get good enough that most companies will eventually stop hiring someone that doesn’t utilize AI. Because most companies are just making crud (pun intended). It’ll be like specialized programming languages. Some will exist, and they may get paid a lot more, but most people won’t fall into that category. As much as we like to puff ourselves up, our profession isn’t really that hard. There are a relative handful of people doing some really cool, novel things. Some larger number doing some cool things that aren’t really novel, just done very nicely. And the majority of programmers are the rest of us. We are not special.
What I don’t know is the timing. I don’t expect it to be within 5 years (though I think it will _start_ in that time), but I do expect it within my career.
aeon_ai
AI isn't a stylistic preference or minor enhancement, but cognitive augmentation that allows developers to navigate complexity at scales human cognition wasn't designed for.
Just as the developer who refused to adopt version control, IDEs, or Stack Overflow eventually became unemployable, those who reject tools that fundamentally expand their problem-solving capacity will find themselves unable to compete with those who can architect solutions across larger possibility spaces on smaller teams.
Will it be used for absolutely every problem? No - There are clearly places where humans are needed.
But rejecting the enormous impact this will have on the workforce is trading hype goggles for a bucket of sand.
kibwen
> Just as the developer who refused to adopt version control, IDEs, or Stack Overflow eventually became unemployable
This passage forces me to concluse that this comment is sarcasm. Neither IDEs nor the use of Stack Overflow is anywhere near a requirement for being a professional programmer. Surely you realize there are people out there who are happily employed while still using stock Vim or Emacs? Surely you realize there are people out there who solve problems simply by reading the docs and thinking deeply rather than asking SO?
The usage of LLM assistance will not become a requirement for employment, at least not for talented programmers. A company gating on the use of LLMs would be preposterously self-defeating.
jraph
> cognitive augmentation that allows developers to navigate complexity at scales human cognition wasn't designed for
I don't think you should use LLMs for something you can't master without.
> will find themselves unable to compete
I'd wait a bit more before concluding so affirmatively. The AI bubble would very much like us to believe this, but we don't yet know very well the long term effects of using LLMs on code, both for the project and for the developer, and we don't even know how available and in which conditions the LLMs will be in a few months as evidenced by this HN post. That's not a very solid basis to build on.
aeon_ai
Two masters go head to head. One uses AI tools (wisely - after all, they're a master!), the other refuses to. Which one wins?
To your second point -- With as much capital as is going into data center buildout, the increasing availability of local coding LLMs that near the performance of today's closed models, and the continued innovation on both open/closed models, you're going to hang your hat on the possibility that LLMs are going to be only available in a 'limited or degraded' state?
I think we simply don't have similar mental models for predicting the future.
diggan
> I don't think you should use LLMs for something you can't master without.
I'm not sure, I frequently use LLMs for well-scoped math-heavy functions (mostly for game development) where I don't neccessarly understand what's going on inside the function, but I know what output I expect given some inputs, so it's easy for me to kind of blackbox test it with unit tests and iterate on the "magic" inside with an LLM.
I guess if I really stopped and focused on math for a year or two I'd be able to code that myself too, but every time I tried to get deeper into math it's either way too complex for me to feel like it's time well spent, and it's also boring. So why bother?
laughingcurve
“I don’t think you should pay others for something you can’t master without [paying].” is one a hell of an argument to make. Good luck trying.
jennyholzer
IMO Version Control, IDEs, and Stack Overflow are many degrees of magnitude more valuable than GPT tools.
The use cases of these GPT tools are extremely limited. They demo well and are quite useful for highly documented workflows (E.G. they are very good at creating basic HTML/JS layouts and functionality).
However, even the most advanced GPT tools fall flat on their face when you start working with any sort of bleeding edge, or even just less-ubiquitous technology.
philipp-gayret
That is interesting, in my experience these tools work quite well on larger codebases but it depends how you use them and I haven't really found a counterexample. Do you maybe have a practical example, like a repo you could link that just doesn't work for AI?
n4r9
This may not be the strong argument you think it is. There are plenty of highly productive senior developers who either don't use IDEs or SO or use them very minimally. Even version control, if they're working alone. Smart devs will find out how to be virtually as productive in a terminal as they would be with an IDE. Potentially more productive when solving edge case issues with processes that IDEs abstract
oblio
IDEs can be iffy, but any project bigger than a 20 line throwaway script needs/deserves version control.
contravariant
If reading your code requires navigating complexity that human cognition wasn't designed for then something has gone terribly wrong.
conartist6
Really, scales human cognition wasn't designed for?
Human cognition wasn't designed to make rockets or AIs, but we went to the moon and the LLMs are here. Thinking and working and building communities and philosophies and trust and math and computation and educational institutions and laws and even Sci Fi shows is how we do
Chris2048
> we went to the moon
We also killed quite a few astronauts.
drw85
I think this is a somewhat short sighted perspective. It's not really augmenting, but replacing cognition.
I see people starting to unlearn working by themselves rapidly and becoming dependant on GPT, making themselves quite useless in the process. They no longer understand what they're working with and need the help from the tool to work. They're also entirely helpless when whatever 'AI' tool they use can't fix their problem.
This makes them both more replaceable and less marketable than before.
It will have and already has a huge impact. But it's kinda like the offshoring hype from a decade ago. Everyone moved their dev departments to a cheaper country, only to later realize that maybe cheap does not always mean better or even good. And it comes with a short term gain and a long term loss.
catmanjan
Maybe somehow this will be true in the future, but I am finding that as soon as you work on a novel or undocumented or non internet available problem it is just a hallucinating junior dev
mcny
The dirty secret is most of the time we are NOT working on anything novel at all. It is pretty much a CRUD application and it is pretty much a master detail flow.
mark_l_watson
+1, even though I mildly disagree with you. I pay For Gemini Pro by the year, and even though I don’t use it often, it is still high value. There are obvious things like generating a Bash shell script quickly - and other things I rarely do, are simple, and I save 5 minutes here and there. Sometimes code generation can be useful, in moderation.
But the big thing is using AI to learn new things, explain some tricky math in a paper I am reading, help brain storm, etc. The value of AI is in improving ourselves.
jennyholzer
> explain some tricky math in a paper I am reading
To me this seems to be the single most valuable use case of newer "AI tools"
> generating a Bash shell script quickly
I do this very often, and to me this seems to me the second most valuable use case of newer "AI tools"
> The value of AI is in improving ourselves
I agree completely.
> help brain storm
This strikes me as very concerning. In my experience, AI brainstorming ideas are exceptionally dull and uninspired. People who have shared ideas from AI brainstorming sessions with me have OVERWHELMINGLY come across as AI brained dullards who are unable to think for themselves.
What I'm trying to say is that Chat GPT and similar tools are much better suited for interacting with closed systems with strict logical constraints, than they are for idea generation or writing in a natural language.
mark_l_watson
For brainstorming: when I write out a plan for a writing or coding project, I like to ask questions for ‘what am I missing or leaving out?’, etc.
Really, it is like students using AI: some are lazy and expect it to do all the work, some just use it as a tool as appropriate. Hopefully I am not misunderstanding you and others here, but I think you are mainly complaining about lazy use of AI.
Kim_Bruning
A surprisingly large number of musicians do use things like written music (a medieval invention, but came into its own in the renaissance) or amplifiers (a modern one).
conartist6
Sure but that's different than saying "if you don't use those things you are worthless and will fail"
Kim_Bruning
Ah, I personally read it as "Use of good tools makes you more competitive",
but you're right that "I firmly believe that AI will not replace developers, but a developer using AI will replace a developer who does not." could have multiple other readings too.
shepherdjerred
This is true for 99% of developers who aren't particularly talented or driven, e.g. the average engineer who treats their job as a job and not their passion.
jennyholzer
To be completely honest I feel that developers who use AI are beneath me.
victorbjorklund
I'm sure that is how many photographers felt about digital cameras and photo editing software when it first came. Now very few professional photographers work without digital cameras and photo editing software (even if it is just to tweak some colors). Yes, you will still have some artists making money with purely analog cameras without any touch up but they are a tiny minority of the photographers getting paid for their work (slightly more people use analog cameras as a hobby).
boleary-gl
It's probably true in our industry too - I'm sure that when people started using programming languages instead of writing machine code people looked down on them too.
This will be another abstraction layer that MANY people will use and be able to accomplish things that would have been impossible to do in a reasonable amount of time in machine code.
jennyholzer
I'm sure that's how many laundry machine users felt when the tide pod first came out.
Now very few laundry doers measure out their detergent by hand.
daveguy
> I'm sure that is how many photographers felt about digital cameras and photo editing software when it first came.
And for a few decades at least it was true. The technology was shite compared to film photography for a long time. The same will probably be true for AI, as full developer replacement will require AGI.
TrackerFF
I think it is great to take pride in ones craft, but in the end it comes down to this: We create software for users.
If the user can't feel any difference in quality between human made software, or AI made software, then it does not mater. It is that easy.
If AI makes better software, at lower prices, human developers will become obsolete. It is the natural way of life.
Once we had telephone operators, now we don't. Once you had to be a good tradesman with good knowledge in how to use a mallet/hammer, axe, chisel, etc. to build a house - now you don't. You don't get awarded for being old-fashioned.
jennyholzer
I make better software because I don't think like this.
theshrike79
There's still work for artisanal woodworkers, doing everything by hand with the utmost care[0]
But that doesn't change the fact that the VAST majority of people are just fine with mass-produced furniture
I think this is the difference that's going to happen to software.
There's the one doing everything in bare vim with zero assist, just rawdogging function names and library methods from rote memory.
And then there's the rest who use all the tools at their disposal to solve problems. Is the work super clean, efficient and fully hand-crafted? Nope. Does it solve the problem? Yes it does, and fast.
baq
Woodworkers who use powertools are beneath artisans with only hammers and chisels, too, right?
Current gen of LLMs are tools. Use them or not. Judging others based on whether they use tools at all vs how they use them is… naive.
alt227
Typical opinion of someone who has not yet tried developing with ai.
When you finally give it a go you will feel stupid for having this opinion.
I did.
jraph
I've seen a very clever dev use it. I can't know if you are even cleverer, but I'd be cautious.
In any case there are better and stronger arguments against LLMs than this.
shepherdjerred
Such a strong reaction sounds much more driven by emotion than reason
tiahura
do you hand compile your code?
TrackerFF
When you offer "unlimited", you're immediately exposing yourself to the 0.1% of userbase that will absolutely treat it is unlimited. Whale users/hoarders. This has been a known problem since the dawn of hosting...hell, it goes back to times before computers.
One thing I miss for the other users, i.e. the casual users that never use anywhere near of their quota, is rollover. If you haven't used your quota this month, the unused will roll over to the next month.
thimabi
Things would be much easier if, instead of “unlimited”, companies simply offered great usage limits. Yes, hoarders would still exist, but their impact would be negligible.
Even better: provide a counter displaying both remaining usage available and the quota reset time.
But companies probably earn so much money from the vast majority of users that having good and clear limits would only empower them to actually benefit as much from the product as they can.
jollyllama
It simply shouldn't be called unlimited, then.
const_cast
This is the crux of the issue.
Every company wants the marketing of unlimited, but none of them want the accountability.
vertoc
I think the hard part with these ai models is it’s kind of hard to figure out how many tokens you’re going to use, especially as a new user. 4,000 tokens sounds like a lot for instance, but is tiny
thimabi
If companies offered any sort of counter, users would quickly understand their usage patterns and adapt accordingly. But users are kept in the dark on purpose, often not knowing how many tokens they use, and sometimes not even knowing what are the token limits.
fuzzzerd
While I don't disagree that determining usage is incredibly difficult, it doesn't take a genius to see that something sold "by the million" is something you can expect to to use a lot of.
ollybee
Two problem with offering usage limits is the real limit you could offer if all users hit it, is low. The users with usage far below the limits feel they are getting a bad deal, compared to if they can't see the limits and they don't hit them, they feel they have "unlimited!".
Spooky23
I disagree. People will always hoard a constrained resource. Things like phone service and AI tools find their value when people use them freely.
The AI models have a bunch of different consumption models aimed at different types of use. I work at a huge company, and we’re experimenting with different ways of using LLMs for users based on different compliance and business needs. The people using all you can eat products like NotebookLM, Gemini, ChatGPT use them much more on average and do more varied tasks. There is a significant gap between low/normal/high users.
People using an interface to a metered API, which offers a defined LLM experience consume fewer resources and perform more narrowly scoped tasks.
The cost is similar and satisfaction is about the same.
OldfieldFund
An even more insidious method is selling "lifetime access," offering incredible over-the-top value, then exiting after 2 or 3 years when the momentum starts dropping. These are essentially rug pulls/exit scams.
There is no such thing as "unlimited" or "lifetime" unless it's self-hosted.
null
null
windward
>This has been a known problem since the dawn of hosting...hell, it goes back to times before computers.
yep
>Adverse selection has been discussed for life insurance since the 1860s,[3] and the phrase has been used since the 1870s.[4]
Aurornis
You’re fully right and I’ve seen it play out at a startup. It’s unreal to see how that 0.1% userbase can find insane ways to use your service. For some people, it becomes a game to discover your rate limits and develop a service that goes just under the rate limit 24/7
In some cases, people discover creative ways to resell the service. Anthropic mentioned they suspect this was happening.
The weirdest part about this whole internet uproar, though, is that Anthropic never offered unlimited usage. It was always advertised as higher limits.
Yet all the comment threads about it are convinced it was unlimited and now it’s not. It’s weird how the internet will wrap a narrative around a story like this.
JumpCrisscross
> 0.1% of userbase that will absolutely treat it is unlimited. Whale users/hoarders
This is somewhat a different issue that’s largely accepted by courts and society bar that one neighbour who is incensed they can’t run a rack off their home internet that was marketed unlimited.
CraigRood
Thing is, even with users that don't use the quota, these AI companies are still losing money. This isn't a case of the small users paying for large. The true costs of AI are yet to unravel.
SirFatty
Like the unlimited meal pass at Six Flags.. https://www.vice.com/en/article/meet-the-guy-whos-finessing-...
Or the American Airlines lifetime pass.. https://www.aerotime.aero/articles/american-airlines-unlimit...
dcminter
One does find it occasionally. My mobile phone plan (in Sweden) currently says that I have 774G left of my 25G/month quota for example :)
Philpax
Thank you, Hallon - I've been on the same plan for the last four years and I don't think I've ever checked my data quota :)
johnisgood
So it accumulates for you? That is nice. I wish my unused quota would go into the next month, but it resets. :(
kqr
One might wonder why you pay for a plan with 25 GB/mo when you use so little that you have over 2.5 years of it saved!
I thought I had a low usage with my 1.5 years' worth saved. Only reason I paysfor that plan is anything lower and my provider does not offer rollover.
dcminter
Because it's inexpensive and I'd rather pay marginally over the odds than ever have to think about it (I'm not even sure there was a cheaper plan).
ajsnigrutin
Sometimes these are the smallest usable packages.
Eg here in slovenia, if you want unlimited calls and texting, you get 150GB in your "package" for 9.99eur, but you somehow can't save that data for the next month.
https://www.hot.si/ponudba/paketi.html (not affiliated)
blitzar
Data limits are easier to keep "unlimited" by slowing the speed, which was already finite, to barely servicable levels. If it is clearly laid out this seems a reasonable solution.
dcminter
If they've ever done that then I've never noticed it - and mine doesn't claim to be unlimited.
raverbashing
Still limited. But "unlimited" to most people
In the same way your next-door supermarket has effectively "infinite soup cans" for the needs of most people.
wat10000
Unlimited means for a flat rate. Pay per use/item, like at the supermarket, isn’t it.
null
sasmithjr
I didn't look at the URL at first and was surprised when this turned in to an ad. Oh well!
> Stop selling "unlimited", when you mean "until we change our minds"
The limits don't go in to affect until August 28th, one month from yesterday. Is there an option to buy the Max plan yearly up front? I honestly don't know; I'm on the monthly plan. If there isn't a yearly purchase option, no one is buying unlimited and then getting bait-and-switched without enough time for them to cancel their sub if they don't like the new limits.
> A Different Approach: More AI for Less Money
I think it's really funny that the "different approach" is a limited time offer for credits that expire.
I don't like that the Claude Max limits are opaque, but if I really need pay-per-use, I can always switch to the API. And I'd bet I still get >$200 in API-equivalents from Claude Code once the limits are in place. If not? I'll happily switch somewhere else.
And on the "happily switch somewhere else", I find the "build user dependency" point pretty funny. Yes, I have a few hooks and subagents defined for Claude Code, but I have zero hard dependency on anything Anthropic produces. If another model/tool comes out tomorrow that's better than Claude Code for what I do, I'm jumping ship without a second thought.
theshrike79
TBH paying yearly for ANY LLM tool at this time is just pure insanity.
The field is moving so fast that whatever was best 6 months ago is completely outdated.
And what is top tier today, might be trash in a few months.
piker
The second sentence is revealing. The creation of the 787 didn’t make the 747 “trash”.
mentalgear
E.g.: Apple's new "iCare" offers "unlimited" repairs by subscribing for 20 USD / month!
- note: "unlimited" does not mean free.
quote source: "Apple Just Found a Way to Sell You Nothing" https://www.youtube.com/watch?v=ytkk5NFZGjs
Aurornis
> - note: "unlimited" does not mean free.
Repairs have always come with deductibles.
This is standard in virtually every insurance program. There are a lot of studies showing that even the tiniest amount of cost sharing completely changes how people use a service.
When something is unlimited and free, it enticed people to abuse it in absurd ways. With hardware, you would get people intentionally damaging their gear to get new versions for free because they know it costs them nothing.
diabllicseagull
sounds a lot like any other insurance where the next part of the game is finding reasons why they can't provide the promised repairs.
https://www.forbes.com/sites/barrycollins/2024/11/28/mac-own...
benterix
That's crazy, I misinterpreted just like others, that you pay a monthly but repairs are free. But on re-reading the terms, it turns out this is just pure bullshit. Why would anyone buy this "new iCare"?!
tonyhart7
apple fanboy/fangirl
jajko
Pretty typical of Apple, and the reason for so is its users, whole world knows that Apple can charge premium and sometimes deliver it, sometimes cheaper competition is better or more durable.
Don't blame the company, it acts within boundaries allowed by its paying customers, and apple customers are known to be... much less critical of the company and its products to be polite, especially given its premium prices.
philistine
> and apple customers are known to be... much less critical of the company
This is patently false and has been for the whole existence of Apple. Apple customers are voraciously critical of the company. Just probably not under the delta of importance that you consider.
sudhirj
Did the Max plan ever promise unlimited anything? I’m on it, and I remember seeing and paying for 20x, not infinity.
There is a case to be made that they sold a multiple and are changing x or rate limiting x differently, but the tone seems different from that.
davidbarker
No, it's never promised unlimited — it's always had usage limits: 20× the usage of their regular Pro plan, with a limit of 50 sessions per month (a session being a 5-hour window), although I don't know if they ever enforced this.
They appear to have removed reference to this 50-session cap in their usage documents. (https://gist.github.com/eonist/5ac2fd483cf91a6e6e5ef33cfbd1e...)
So even if these mystery people Anthropic reference who did run it "in the background, 24/7", they still would've had to stay within usage limits.
mrits
Looking at their pricing page in way back machine looks like they've had "usage limits" terminology for at least the last year
Aurornis
I can’t find any indication it was ever sold as unlimited.
It always had limits and those limits were not specified as concrete numbers.
It’s amazing how much of the internet outrage is based on the idea that it was unlimited and now it’s not. The main HN thread yesterday was full of comments complaining about losing unlimited access.
It’s so weird to watch people get angry about thinking they’re losing something they never had. Even Anthropic said less than 5% of accounts would even notice the new limits, yet I’ve seen countless comments raging that “everyone must suffer” due to the actions of a few abusing the system.
bananapub
> Did the Max plan ever promise unlimited anything?
no, even their announcement blog[0] said:
> With up to 20x higher usage limits
in the third paragraph.
mattlondon
I feel like this is where the heavyweights are really going to start throwing their weight around and basically drive the smaller startups out of town.
Can you really ever compete when you are renting someone else's GPUs?
Can you really ever compete when you are going up against custom silicon built and deployed at scale to run inference at scale (i.e. TPUs built to run Gemini and deployed by the tens-of-thousands in data centers around the globe)?
Meta and Google have deep pockets and massive existing world-class infrastructure (at least for Google, Meta probably runs their php Facebook thing on a few VPS dotted around in some random colos /s ) . They've literally written the book on this.
It remains to be seen how much more money OpenAI can burn, but we've started to see how much Anthropic can burn if nothing else.
superasn
I'm not a fan of usage caps either, but that Reddit post [1] (“You deserve harsh limits”) does highlight a perspective worth considering.
When some users burn massive amounts of compute just to climb leaderboards or farm karma, it’s not hard to imagine why providers might respond with tighter limits—not because it's ideal, but because that kind of behavior makes platforms harder to sustain and less accessible for everyone else. Because on the other hand a lot of genuine customers are canceling because they get API overload message after paying $200.
I still think caps are frustrating and often too blunt, but posts like that make it easier to see where the pressure might be coming from.
[1] https://www.reddit.com/r/ClaudeAI/comments/1lqrbnc/you_deser...
apwell23
lets not blame shift anthropic bait and switch to 'bad users' .
Surely they thought about 'bad users' when they released this product. They can't be that naive.
Now that they have captured developer mindshare. users are bad.
bananapub
> anthropic bait and switch
what was the bait and switch? where in the launch announcement (https://www.anthropic.com/news/max-plan) did they suggest it provided unlimited inference?
apwell23
so where did 'bad users' come from if users were simply doing what they were allowed to?
why is anthropic tweeting about 'naughty users that ruined it for everyone' ?
mmillin
Did Anthropic ever use the term unlimited? I understand the general frustration with the pattern, but it seems weird to put unlimited in quotes when it wasn’t the way Claude was sold.
seunosewa
Nope.
cedws
There’s only one way unlimited plans work: either the seller gets scammed, or the buyer does.
When companies sell unlimited plans, they’re making a bet that the average usage across all of those plans will be low enough to turn a profit.
These people “abusing” the plan are well within their right to use the API as much as they want. It just didn’t fall into the parameters Anthropic had expected.
LLM subscriptions need to go away, why can’t we just pay as we go? It’s the fairest way for everyone.
Aurornis
> When companies sell unlimited plans,
Anthropic never sold an unlimited plan
It’s amazing that so many people think there was an unlimited plan. There was not an unlimited plan.
> These people “abusing” the plan are well within their right to use the API as much as they want. It just didn’t fall into the parameters Anthropic had expected.
Correct! And they did. And now Anthropic is changing those limits in a month.
> LLM subscriptions need to go away, why can’t we just pay as we go? It’s the fairest way for everyone.
This exists. You use the API. It has always been an option. Again, I’m confused about why there’s so much anger about something that already exists.
The subscriptions are nice for people who want a consistent fee and they get the advantage of a better deal for occasional heavy usage.
cedws
>Anthropic never sold an unlimited plan
I'm told the $200/month plan was practically unlimited, I heard you could leave ~10 instances of Claude Code running 24/7. I will never pay for any of these subscriptions however so I haven't verified that.
>And now Anthropic is changing those limits in a month.
Which indicates the seller was being scammed. Now they're changing the limits so it swings back to being a scam for the user.
>I’m confused about why there’s so much anger about something that already exists
Yes but much LLM tooling requires a subscription. I'm not talking only about Anthropic/Claude Code. I can't use chatgpt.com using my own API key. Even though behind the scenes, if I had a subscription, it would be calling out to the exact same API.
redhale
Nothing is stopping you from using the API directly, if you prefer to donate more money to Anthropic.
I would not personally, as I can't spend thousands per month on an agentic tool. I hope they figure out limits that work. $100 / $200 is still a great deal. And the predictability means my company will pay for it.
senko
I am using Claude daily, exclusively via the API (in Zed, added my own token) and spend a few bucks a day tops.
Unlimited plans encourage wasting resources[0]. By actually paying for what you use, you can be a bit more economical and still get a lot of mileage out of it.
$100/$200 is still a great deal (as you said), but it does make sense for actually-$2000 users to get charged differently.
0: In my hometown, (some) people have unlimited central heating (in winter) for a fixed fee. On warmer days, people are known to open windows instead of turning off the heating. It's free, who cares...
Aeolun
> LLM subscriptions need to go away, why can’t we just pay as we go? It’s the fairest way for everyone.
Because Claude Code is absolutely impossible to use without a subscription? I’m fine with being limited, but I’m not with having to pay more than $200/month
Anybody that feels they’re not getting enough out of their subscription is welcome to use API instead.
Aurornis
> Because Claude Code is absolutely impossible to use without a subscription?
Claude Code accepts an API key. You do not need a subscription
https://docs.anthropic.com/en/docs/claude-code/settings#envi...
Aeolun
I mean the way it works quickly leads to exorbitant costs. Subscription based pricing with limits keeps everyone honest.
aeon_ai
The bait-and-switch critique is valid, but the real pragmatic issue is that AI companies are discovering their unit economics don't support flat-rate pricing for compute-intensive services. Try running AWS on a Netflix subscription model.
The transparency problem compounds this. The sustainable path forward likely involves either much more transparent/clear usage-based pricing or significantly higher flat rates that actually cover heavy usage.
cadamsdotcom
In Australia they outlawed advertising “unlimited” broadband, because it was a deceptive term. Providers actually meant “slower than a modem after a certain amount of download”.
Everyone was better off without the deception. Now we are in the early days of AI. Providers should be honest but won’t until forced to.
Because just think about it. Unlimited is untenable. Another example, in the early days of broadband in Australia a friend’s parents were visited by a Telstra manager because he “downloaded more than his entire suburb”. A manager!
Really you can’t blame the providers; some users will ruin it for everyone. I am not saying that is anyone specific. But none of this should surprise us. We’ve been here before. Just look back at how other markets developed & you will see patterns that tell you what’s next.
I'm trying to cancel my Claude plan right now, because mid-research I got hit with a one hour time out for the first time, and I realized just how little I need the "expensive" research stuff and just want basic language refinement and some basic augmented knowledge search to get more terms to search the web. And all that basic stuff gets cutoff when you hit some invisible barrier that you have no way of controlling or monitoring.
But I literally can not cancel. Trying the app says "you signed up on a different platform, go there" but it doesn't tell me which platform that might be.
Trying to cancel on mobile web gives several upgrade options but no cancel options.
So, do I need to call my credit card? This is the worst dark pattern on subscription I have seen of any service I have ever paid for!
Anthropic had a fairly positive image in my head until they cut off my access and are not giving me a way to cancel my plan.
Edit: after mucking with the Stripe credit card payment options I found a cancel plan button underneath the list of all invoices. So there is an option, I just had a harder time finding it then I have had with other services. Successfully cancelled!