> If you can find an LLM+TTS generated podcast with even a FRACTION of the infectious energy

this is the wrong standard. there's a long tail of many many many many topics that a Radio Lab will never have the resources or critical mass or staff or turnaround time to cover. AI can serve the long tail like nothing you have ever seen. (because humans are expensive, and deservedly so, but our needs and wants are infinite)

You're not wrong, but your argument seems to boil down to "something is better than nothing" and I suppose I disagree. If there's not a podcast about a particular topic that I'm interested in, I would rather go find a curated expose of that topic via a different medium, such as longform journalism.

I think you make a good connection with the long tail. That will definitely be one of AI's triumphs. People who don't understand this, don't understand what's been happening to media production and consumption over the last 40 years.