I believe all the AI providers are reaching capacity limits, and they want to preserve their margin for customers paying by the token, instead of monthly plans(which are effectively subsidize advertising).

It won't just be Anthropic.

> From a naive perspective, more usage should mean more revenue since customers pay per token.

They are not limiting usage through the API with per token payment.

They are limiting usage through a subscription that gives you a lot (but they won't tell you how much exactly) of tokens for a fixed monthly payment.

Using models through a subscription tends to be much cheaper than using the API and paying per token.

Anthropic provides cheap subscriptions with less profit or even at a loss (we don't know) to promote its own tools. Using subscription for third-party tools obviously defeats this purpose.

Yeah but what if you sold shoe subscriptions because mostly people only wear one pair of shoes at a time -- but some people were using a service called OpenToe that was grabbing 50-100 shoes at a time?

If you’re selling shoes to potentially 10,000 people, but ten customers consistently buy your entire stock, then you’re making the same money. But you’re limiting your future expansion, because whenever the other 9,990 potential customers try to buy your shoes they find none on the shelves and either choose to not use you in the future, or worse tell people how bad you are at supplying shoes.

Given finite server availability, my guess is that OpenClaw users were degrading the service (or start hitting service limits soon) for “normal” users, causing Anthropic to get worried that their cashcow in Claude would start to get negative press.

They operate at loss.

I think the strategy you're suggesting is: "We lose money on every sale, but make it up in volume!"

If the resellers down the chain were purchasing your shoes for less then your cost, would you still be happy?

Say the resellers were abusing an 80% discount coupon. Anthropic is basically closing a 95% discount coupon that was being abused.

If OpenClaw users were paying the API rate, your strategy could make more sense.

The reason Anthropic is subsidizing inference is because they are trying to capture users (marketshare). However the acquisition costs for a single OpenClaw user is much higher. And OpenClaw users are less likely to convert into profitable users later.

---

In addition, there is a supply bottleneck. Currently Anthropic is having trouble servicing all the demand due to a shortage of GPU's. And in the current market it is impossible to get more GPU's (or at least prohibitively expensive).

Anthropic (and all other AI companies) also need GPU's to stay competitive: GPU's are needed to train better models. So you could view it as Anthropic has decided instead of subsidizing nonprofitable OpenClaw users, it is better to repurpose that GPU for internal R&D instead.