On: Are OpenAI and Anthropic losing money on inference?

[doctorpangloss]

I’m pretty sure input tokens are cheap because they want to ingest the data for training later no? They want huge contexts to slice up.

[awwaiid → doctorpangloss]

Afaik all the large providers flipped the default to contractually NOT train on your data. So no, training data context size is not a factor.