I agree. I think it's more of a niche use-case than image models (and fundamentally harder to evaluate), but transcription and summarization is my current front

I agree. I think it's more of a niche use-case than image models (and fundamentally harder to evaluate), but transcription and summarization is my current front-runner for winning use-case of LLMs. That said, "hallucination" is more of a fundamental problem for this area than it is for imagery, which is why I still think imagery is the most interesting category.