There is not doubt that podcasts are on the rise. According to Backlinko:
🔹Over 584.1 million podcast listeners worldwide
🔹55% of the US population 12+ listen to a podcast at least once a month
🔹Podcast ad spending is estimated to hit $4.46 billion in 2025
So I wondered if AI engines were indexing the audio. I asked that question to ChatGPT, Gemini, CoPilot, Claude and Perplexity. Of those AI engines, ONLY Google was able to crawl audio files.
Google was the standout because it has has its own speech-to-text (STT) technology called Google Cloud Speech-to-Text. It is an AI service that was designed to convert audio into text transcriptions. It is also something you can see at work in YouTube.
All of the other AI Engines were NOT able to crawl audio files. Specifically, ChatGPT said:
🗨I don’t pull in raw podcast audio. When I “browse,” I’m limited to text that’s already on the web—so if a podcast episode has a published transcript or detailed show notes, I can read those, but I don’t stream or transcribe the audio myself. My underlying training may include some publicly available podcast transcripts, but not the audio files.🗨
What does that mean? If you are not:
🔹Creating web pages for your podcast
🔹Providing full transcript text
🔹Making your brand / web the owner of that transcript content
Your information is not going to be include in search AI results with the potential exception of Gemini.
