AI Disclosure in Podcasting: The Challenge, The Debate, and What Comes Next

The term "AI slop" is gaining traction in podcasting. Mass-produced, AI-generated shows are flooding directories at industrial scale, with no disclosure to listeners. But AI in podcasting is not all bad. Voice cloning enables multilingual content, post-production tools improve audio quality, and creators use AI as a legitimate production tool every day. The question is not whether AI belongs in podcasting. The question is: should listeners know?

In this episode, we break down the AI disclosure challenge facing the podcast industry. We cover the four major proposals debated on GitHub over the past 18 months within the Podcasting 2.0 namespace, including contributions from developers, app builders, and industry voices. We look at what companies like Spreaker and RSS.com have shipped so far, why writing disclosure to the open RSS feed matters, and what Apple's Content Guidelines already require from podcasters.

This podcast episode uses an AI generated voice, and its content was also generated by AI based on public sources (so it may contain errors). This is also disclosed in the RSS feed using the Podcasting 2.0 podcast:txt tag.

Sources referenced: Podnews, GitHub Podcasting 2.0 namespace (Discussions #663, #669, #731, #735), Apple Podcasts Content Guidelines, public Mastodon posts.

More resources:

• AI Disclosure in Podcasting: What It Is, Why It Matters, and How to Do It
• AI Disclosure in Podcasting: a clear, actionable set of guidelines for podcasters
• The Substance Test: a quick questionnaire to help you decide whether to disclose AI