Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Anthropic's best AI model struggles to attract users as cheaper tools thrive

ft.com

Post reply on HN
Print view View original

521–530 of 740 posts

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#521

Post by suddenlybananas »

nrmitchi wrote: post #399
hirvi74 wrote: post #314

Earlier quoted context omitted.

I have a family member that is an attorney in housing law. She claimed that LLMs are not particularly useful for her work. If she asked a simple question like, "Find all the for all 50 states," then she still has to go and check every single one of the laws. Since the legislature is modified so often, she cannot look at, say, Maryland's law and know if it the LLM output was the 1990, 2014, 2018, or 2026 version of th…

This doesn’t sound as much that “LLMs can’t help me” as much as “trying to one-shot a complicated answer in a chatbot can’t help me”. Output straight from a generative model definitely should not be trusted. But that doesn’t mean that an appropriate harness (even Claude code and a dynamic workflow) can’t complete, and validate, an answer.

Seems like it would be a massive market to make such a harness, and yet you're making it sound very easy to do. It would be surprising no one has done it if it were.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#522

Post by lelanthran »

visarga wrote: post #330
a1371 wrote: post #113

Where Anthropic f'ed up was treating their monetization the way they treat model training. Turns out that success in experimentation is not transferrable. They have tried to find the highest that the market pays for sota models; however, on the consumer side, this is just too confusing and unsettling: "You can only use Fable for a week as a part of your plan" "Be ready! You have to start paying per token!" "Nevermind…

> That forces people to look beyond the walled garden. Every time I get "you used your quota, come back in 3 hours, or 2 days" -> that is experimentation time with their competition, leading to changed service plans. When they said "claude -p" will be billed at API pricing even for plan users I moved my harness off claude. After I integrated codex, then it was never going to be a full claude project again. What busin…

> What business encourages users to try their competition and adapt their usage to the competing products?

If you are selling something, and losing $10 on each sale, you also would want to limit how much you sell.

I mean, sure, you are losing money on each sale so you can landgrab, but you still have to balance the land-grabbing with how much money you can actually lose.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#523

Post by benjiro29 »

dyauspitr wrote: post #207
a1371 wrote: post #113

Where Anthropic f'ed up was treating their monetization the way they treat model training. Turns out that success in experimentation is not transferrable. They have tried to find the highest that the market pays for sota models; however, on the consumer side, this is just too confusing and unsettling: "You can only use Fable for a week as a part of your plan" "Be ready! You have to start paying per token!" "Nevermind…

DeepSeek is doing pretty much the same thing they said they weren’t going to change their prices after the 75% discount for the foreseeable future. That foreseeable future turned out to be two months.

It did not exactly help that we saw traffic to DS (over OpenCode) jump from around 1.6T tokens per day, to over 14T token in a matter of days. Nobody has the compute to deal with such increases.

This keeps happening with every good new model release. People jumping from one to another, and as prices get lower, they start using the models even more.

People make not like to hear it but prices and usage limits are ways to shape traffic. The third option is the nuclear one like Kimi did, by just stopping to sell subscription at all. But that is something that DeepSeek can not do as all they offer is API.

Even OpenAI despite having the most compute is not immune to client influx = capacity issues. As people found their usage dropping, despite the push to the easier to run Luna models.

Reality is, that compute can not keep up with demand, especially when models get more capable and cheaper. What trigger people being using them more, what trigger compute crisis's.

This constant up and down cycle is going to keep happening for a long time, as this new market grows and eventually, somewhere in the future stabilizes. But yea, that is still going to be a few more years for sure.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#524

Post by Scaled »

mrtsepelev wrote: post #436
dexterlagan wrote: post #405

Earlier quoted context omitted.

Qwen 3.8 is excellent. With the right harness, it does about 95% of what Opus can do, in my case automation software development. Since 3.8 came out, I have significantly revised my expectations for a local model. Give it another year or two, and we'll be running fast and free local models for nearly everything that matters, and these costly subscriptions will be a thing of the past. I've always believed that AI shou…

What harness would you recommend? I’ve tried Pi but the model struggled to stay on track after the compaction. I have only 48gb of ram, so can fit only 80k context max, so good compaction is must.

Not op, but check out open code; you can turn on K/V quantization to help with increasing context if you have not already. I think K needs to stay at least 8 but I hear V can go down to 4?

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#525

Post by llm_nerd »

watwut wrote: post #391
llm_nerd wrote: post #211

Earlier quoted context omitted.

>You don't feel they want to give you a dependable service for an, albeit premium, price. It's a constant bargaining game We all know that the subscription prices are not at all sustainable for these providers. You all do, right? Yes, they're struggling to segment the market and find a way to make money, and that basically relies upon emptying the pockets of whales. As someone enjoying a hilariously subsidized Max pl…

I dont see why user should care. OK, they are selling at loss so that they can build a monopoly. That is not something positive or good, not something to cheer on.

This discussion does not concern whether the user "cares". And clearly from my comment I don't "care".

But it explains why a nascent, hyper-competitive (I mean, clearly not remotely a monopoly) market has such erratic policies and pricing.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#526

Post by eastbound »

fluidcruft wrote: post #125
a1371 wrote: post #113

Where Anthropic f'ed up was treating their monetization the way they treat model training. Turns out that success in experimentation is not transferrable. They have tried to find the highest that the market pays for sota models; however, on the consumer side, this is just too confusing and unsettling: "You can only use Fable for a week as a part of your plan" "Be ready! You have to start paying per token!" "Nevermind…

Yeah, I agree with this. The constant state of "...will the rug be pulled?!?" does discourage relying on it as a model and building a workflow on it. Anthropic used to just be a reliable thing you could play with. Now it's this constant source of anxiety. It also didn't help that the government yanked it which adds another source of anxiety since OpenAI is on much better terms with the administration and the administ…

> OpenAI is on much better terms with the administration and the administration seems corrupt enough to...

Which tells you everything you need to know about OpenAI.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#527

Post by intended »

rhdunn wrote: post #500
kbrannigan wrote: post #264

Earlier quoted context omitted.

How many big tech companies let you talk to a human to get support. Automation is wonderful to cut cost for them but for the users being unable to get support is a horrible experience. But you cannot go elsewhere because they are the only player in town. How can small companies with 1000x less money able to provide live support, but if you pay 20, 100, 200 dollars for a subscription you dont have a phone number to ca…

Part of it is scale. If you have a small number of clients/users it is possible to provide that support. If you have millions or billions of users then you can't scale the support to handle the support requests, so some form of automation becomes inevitable.

That's not scale, Its profit margins.

If firms had a base degree of customer support they were expected to provide, they would still exist. They would just not be as profitable, but customers would be better off.

I seem to remember there was a time when S/W was also designed with the aim to be easy to use, so that the need for support was reduced. It feels like the lesson learned was to keep costs low, not to ensure users were ok.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#528

Post by llm_nerd »

PeterStuer wrote: post #383
llm_nerd wrote: post #211

Earlier quoted context omitted.

>You don't feel they want to give you a dependable service for an, albeit premium, price. It's a constant bargaining game We all know that the subscription prices are not at all sustainable for these providers. You all do, right? Yes, they're struggling to segment the market and find a way to make money, and that basically relies upon emptying the pockets of whales. As someone enjoying a hilariously subsidized Max pl…

Do you have actual data on the marginal costs of inferencing, or are you using the term 'subsedizing' loosely as in 'discount', meaning a lower price compared to another price for the same offer?

Look up the burn rate for OpenAI and Anthropic. Despite the fact that these companies have managed to hook a lot of F500 companies into paying millions for token-level API pricing, they are bleeding cash at a catastrophic rate. And we know subscription tokens come at about a 1/30th the cost of the API pricing (presuming you use your quota).

Yes, they are most certainly subsidizing the subscription plans. I mean, at least for people who utilize them to the quota.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#529

Post by jgalt212 »

rhdunn wrote: post #500
kbrannigan wrote: post #264

Earlier quoted context omitted.

How many big tech companies let you talk to a human to get support. Automation is wonderful to cut cost for them but for the users being unable to get support is a horrible experience. But you cannot go elsewhere because they are the only player in town. How can small companies with 1000x less money able to provide live support, but if you pay 20, 100, 200 dollars for a subscription you dont have a phone number to ca…

Part of it is scale. If you have a small number of clients/users it is possible to provide that support. If you have millions or billions of users then you can't scale the support to handle the support requests, so some form of automation becomes inevitable.

It depends more on revenues per customer.

Re: Anthropic's best AI model struggles to attract users as cheaper tools thrive

#530

Post by bean469 »

theshrike79 wrote: post #432
foxylad wrote: post #301

Non-LLM user here. Why? Apart from the ecological issues, I'm very uncomfortable giving my organisation's crown jewels to {random_internet__corp}. Look at the lengths they go to for training data - 10M for Spirit's call logs? Destroying millions of obscure books to scan them? They make meth-heads look scrupulous. Your prompts, especially if they contain your entire codebase, are _way_ more data-rich than an airline p…

That's when you do the Enterprise deal. I have first hand knowledge of household name companies with billion euro IP who use OpenAI and Athropic products quite liberally. They have WAY more lawyers than I do and their sole job is to keep the IP safe. They wouldn't sign a deal with even a slightest whiff of the IP being used to train anything. And if it happens, the penalty for breach of contract would have so many ze…

> Although the LLM providers state that they will not use your data for training when you sign an enterprise deal

It can be very hard to prove in court that your data was used for training

Post reply on HN