3 ms·

It really is the doordash/uber playbook all over again eh? Sell at a massive loss, gain userbase, then gradually boil the frog by adding fees, removing features

It really is the doordash/uber playbook all over again eh? Sell at a massive loss, gain userbase, then gradually boil the frog by adding fees, removing features, and increasing prices. Except instead of doing this a few years down the line, they're speedrunning the tighten-the-noose phase. Unfortunately the competition is nipping at their heels so there's a good chance this blows up in their faces.

intrasight 5mo ago
There is competition. And there is no moat nor network effect. I don't think it'll blow up in their faces if they provide a product and service that people value which demonstratively they have. But it may not be so lucrative to them or their shareholders.
pigpag 5mo ago
[dead]
Aboutplants 5mo ago
“Unfortunately”? Uh, that’s a good thing
ealready_value 5mo ago
As far as I can tell, it seemed very clear that was the playbook for about a year now. Its been regularly assumed they're selling plans as a major loss-leader because people can "spend" thousands of dollars a months on a plan if they were charged at API rates. I think there's good evidence that even the API rates are sold at a loss. I think its assumed in the LLM model business that the models themselves are not a good moat, the next model by another company is just as likely to be as good as the current model. So companies like Anthropic have to tighten the noose slowly to start recovering their costs. This appears to be one of those steps.
- infecto 5mo ago
  The simpler explanation is probably some mix of marketing and also an expected use from people paying for a plan. The money to be made is not from plans ever. It’s in everyone’s best interest for these companies to accurately oversubscribe plans. Enterprise is where the money is to be made and I don’t feel that pricing has changed much on that end.
  - ealready_value 5mo ago
    I had never thought of it that way, but it seems very likely that Enterprise oversubscribing is in the mix. Which does tie in nicely with this change; if a few devs are using their max plan to programmatically run parts of the business that could break the oversubscribes assumption.
    - infecto 5mo ago
      There are no Enterprise plans though only on demand usage which at worst is charged retail token costs and with volume negotiated token rates.
infecto 5mo ago
I think people are being too generous with these comparisons. Not defending Anthropic but at the same time they are releasing new features and adjusting cost at pretty record speed for a new industry. Uber/doordash were subsidizing cost for what felt like a decade. Anthropic and related companies are adjusting price within months. To me the bigger takeaway is that these business are seeing massive volume in use and figuring out how to price the products accordingly.
- hdndjsbbs 5mo ago
  They have to speedrun boiling the frog because the capital expenditure is insane. Remains to be seen just how fast you can boil a frog before the frog notices
  - infecto 5mo ago
    Disagree. Most businesses of size are going to enterprise agreements which are all on demand rates. Those rates have not been changing other than the underlying cost to the model API rates fluctuating. You could make argument they are secretly using that has the price lever. With volume enterprises can already negotiate lower token rates. I don’t see a boiling the frog situation.
    - parliament32 5mo ago
      They will still need to increase costs for enterprise to be profitable, they're just going to be more greasy about it. Claude 5 will cost 20% more but not be 20% better, more shenanigans with "oh no we had a bug in our cache system :^)", or this gem from the current enterprise pricing page: "Opus 4.7 uses a new tokenizer... may use up to 35% more tokens for the same fixed text".
    - hdndjsbbs 5mo ago
      Anecdotally my work is already pivoting from "use as many tokens as possible" to "try and use a dumber model if you can because the best ones are too expensive".
  - 5mo ago
afavour 5mo ago
I still think/hope/pray the future will be on-device models that don't need constant retraining. That will blow up the existing business model but I think a company could still make good money with a "majority local/remote for the really challenging stuff" model. The problem is that today's AI companies have taken on so much funding that a reasonable, not crazy profit ratio isn't enough for them.
- parliament32 5mo ago
  The future is already pretty much here. Note the recent stories about Chrome adding a local model, not to mention the Googlebook demo (if it works as advertised, there's a 0% chance you could get that kind of latency with a non-local model).
- davidw 5mo ago
  If it continues to be a numbers game - the more resources you throw at it, the better it is - then on-device is always going to be not as good. I guess it might be good enough for some uses? I kind of loathe the move away from a world where we could control our own computers and run our own software on them.
doikor 5mo ago
It was very clear from the beginning purely from how much it costs to train and run the inference. Someone has to pay the 7 trillion (the current projections for the AI datacenter build up)
AlexandrB 5mo ago
How do we tell the doordash/uber playbook from the moviepass playbook? Because the latter would be awful to build your business on.
- parliament32 5mo ago
  Moviepass (afaik) was an attempt at the exact same playbook, it just failed. Anthropic will also fail when the competition is.. near-equivalent-capability DeepSeek/Qwen/Llama on a $1k GPU with a break-even of 5 months of subscription costs. The value is simply not there for what they would need to charge to become profitable.
  - gruez 5mo ago
    >when the competition is.. near-equivalent-capability DeepSeek/Qwen/Llama on a $1k GPU with a break-even of 5 months of subscription costs Lol no. Chinese AIs are definitely not "near-equivalent-capability". The empirical proof is pretty obvious: how many people have you heard talking about using their codex/claude code subscription vs their z.ai or qwen subscription? Moreover even the Chinese models require epic amounts of GPUs to run the full version, eg. https://apxml.com/models/glm-51 https://apxml.com/models/glm-51 needs 1515 GB to run, and that's with a measly 1024 token context. To get it to run on your "$1k GPU" you'd need to quantize it, making it even dumber.
    - parliament32 5mo ago
      Today, sure. But we already see diminishing returns with Claude releases, and we know the open models are closing the gap (~6 months behind according to the benchmarks). And when the pitch is "our models are 5% better but cost $200/mo.. also here's a mountain of restrictions" it just won't make sense anymore. Give it a year or two. I could see the "avoid the hardship of running a local model for $20/mo" angle but Anthropic has shown they have little interest in those customers.
      - gruez 5mo ago
        >and we know the open models are closing the gap (~6 months behind according to the benchmarks). Looking at just the benchmarks might be misleading: https://x.com/scaling01/status/2050616057191072161 https://x.com/scaling01/status/2050616057191072161

intrasight 5mo ago

There is competition. And there is no moat nor network effect. I don't think it'll blow up in their faces if they provide a product and service that people value which demonstratively they have. But it may not be so lucrative to them or their shareholders.

pigpag 5mo ago

[dead]

Aboutplants 5mo ago

“Unfortunately”? Uh, that’s a good thing

ealready_value 5mo ago

As far as I can tell, it seemed very clear that was the playbook for about a year now. Its been regularly assumed they're selling plans as a major loss-leader because people can "spend" thousands of dollars a months on a plan if they were charged at API rates. I think there's good evidence that even the API rates are sold at a loss. I think its assumed in the LLM model business that the models themselves are not a good moat, the next model by another company is just as likely to be as good as the current model. So companies like Anthropic have to tighten the noose slowly to start recovering their costs. This appears to be one of those steps.

infecto 5mo ago
The simpler explanation is probably some mix of marketing and also an expected use from people paying for a plan. The money to be made is not from plans ever. It’s in everyone’s best interest for these companies to accurately oversubscribe plans. Enterprise is where the money is to be made and I don’t feel that pricing has changed much on that end.
- ealready_value 5mo ago
  I had never thought of it that way, but it seems very likely that Enterprise oversubscribing is in the mix. Which does tie in nicely with this change; if a few devs are using their max plan to programmatically run parts of the business that could break the oversubscribes assumption.
  - infecto 5mo ago
    There are no Enterprise plans though only on demand usage which at worst is charged retail token costs and with volume negotiated token rates.

infecto 5mo ago

I think people are being too generous with these comparisons. Not defending Anthropic but at the same time they are releasing new features and adjusting cost at pretty record speed for a new industry. Uber/doordash were subsidizing cost for what felt like a decade. Anthropic and related companies are adjusting price within months. To me the bigger takeaway is that these business are seeing massive volume in use and figuring out how to price the products accordingly.

hdndjsbbs 5mo ago
They have to speedrun boiling the frog because the capital expenditure is insane. Remains to be seen just how fast you can boil a frog before the frog notices
- infecto 5mo ago
  Disagree. Most businesses of size are going to enterprise agreements which are all on demand rates. Those rates have not been changing other than the underlying cost to the model API rates fluctuating. You could make argument they are secretly using that has the price lever. With volume enterprises can already negotiate lower token rates. I don’t see a boiling the frog situation.
  - parliament32 5mo ago
    They will still need to increase costs for enterprise to be profitable, they're just going to be more greasy about it. Claude 5 will cost 20% more but not be 20% better, more shenanigans with "oh no we had a bug in our cache system :^)", or this gem from the current enterprise pricing page: "Opus 4.7 uses a new tokenizer... may use up to 35% more tokens for the same fixed text".
  - hdndjsbbs 5mo ago
    Anecdotally my work is already pivoting from "use as many tokens as possible" to "try and use a dumber model if you can because the best ones are too expensive".
- 5mo ago

afavour 5mo ago

I still think/hope/pray the future will be on-device models that don't need constant retraining. That will blow up the existing business model but I think a company could still make good money with a "majority local/remote for the really challenging stuff" model. The problem is that today's AI companies have taken on so much funding that a reasonable, not crazy profit ratio isn't enough for them.

parliament32 5mo ago
The future is already pretty much here. Note the recent stories about Chrome adding a local model, not to mention the Googlebook demo (if it works as advertised, there's a 0% chance you could get that kind of latency with a non-local model).
davidw 5mo ago
If it continues to be a numbers game - the more resources you throw at it, the better it is - then on-device is always going to be not as good. I guess it might be good enough for some uses? I kind of loathe the move away from a world where we could control our own computers and run our own software on them.

doikor 5mo ago

It was very clear from the beginning purely from how much it costs to train and run the inference. Someone has to pay the 7 trillion (the current projections for the AI datacenter build up)

AlexandrB 5mo ago

How do we tell the doordash/uber playbook from the moviepass playbook? Because the latter would be awful to build your business on.

parliament32 5mo ago
Moviepass (afaik) was an attempt at the exact same playbook, it just failed. Anthropic will also fail when the competition is.. near-equivalent-capability DeepSeek/Qwen/Llama on a $1k GPU with a break-even of 5 months of subscription costs. The value is simply not there for what they would need to charge to become profitable.
- gruez 5mo ago
  >when the competition is.. near-equivalent-capability DeepSeek/Qwen/Llama on a $1k GPU with a break-even of 5 months of subscription costs Lol no. Chinese AIs are definitely not "near-equivalent-capability". The empirical proof is pretty obvious: how many people have you heard talking about using their codex/claude code subscription vs their z.ai or qwen subscription? Moreover even the Chinese models require epic amounts of GPUs to run the full version, eg. https://apxml.com/models/glm-51 https://apxml.com/models/glm-51 needs 1515 GB to run, and that's with a measly 1024 token context. To get it to run on your "$1k GPU" you'd need to quantize it, making it even dumber.
  - parliament32 5mo ago
    Today, sure. But we already see diminishing returns with Claude releases, and we know the open models are closing the gap (~6 months behind according to the benchmarks). And when the pitch is "our models are 5% better but cost $200/mo.. also here's a mountain of restrictions" it just won't make sense anymore. Give it a year or two. I could see the "avoid the hardship of running a local model for $20/mo" angle but Anthropic has shown they have little interest in those customers.
    - gruez 5mo ago
      >and we know the open models are closing the gap (~6 months behind according to the benchmarks). Looking at just the benchmarks might be misleading: https://x.com/scaling01/status/2050616057191072161 https://x.com/scaling01/status/2050616057191072161