Depreciation is part of the cost of inference. Inference happens in GPUs that have a relatively short lifespan.

Those GPUs are very expensive.

Inference is expensive because a GPU can only process a certain amount of requests in a given timeframe. Remember that Anthropic is constrained in compute.

If they are constrained, it means that those GPUs are not idle. If they have more customers, they will need more GPUs.

If they have to play silly games using EBITDA to be "profitable", then it means that they need to ramp up prices a lot more than they already did.

Which is why in these discussions I always say that inference is also extremely expensive. Too many people like to pretend without any evidence that inference is cheap.

Anthropic and OpenAI don't own data centers. Since they're renting GPU's, that's not depreciation. Paying rent is an operating cost.

Language models don't wear out the same way; upgrading is a choice.