Spotify teams up with NotebookLM to turn your Wrapped into an AI podcast

Every December, social feeds get hijacked by colorful Spotify Wrapped graphic cards. This year, you don't just have to read about your listening habits—you can listen to two AI hosts roast and celebrate them in real time.

Google announced a partnership with Spotify to bring an official AI podcast feature to the 2024 Spotify Wrapped experience, powered by Google's NotebookLM.

Two AI hosts unpacking your music taste

The feature brings NotebookLM's viral Audio Overviews technology directly into the world's biggest music streaming platform.

Instead of turning PDFs and lecture notes into podcasts, the two AI co-hosts analyze your listening data from the past year. They dive into your top artists, dissect your favorite tracks, and track how your taste shifted across 2024.

Every single podcast is dynamically generated and unique to the listener. Once your episode is ready, you can listen through the Wrapped feed or at spotify.com/wrappedAIpodcast, and download or share it with friends.

There are a few catches. The feature is available only in English and for a limited time. It is restricted to eligible Free and Premium users in the U.S., U.K., Australia, New Zealand, Canada, Ireland, and Sweden.

Google also noted that the AI hosts might occasionally mispronounce names or miss a few details in your listening history.

Why it matters

We think this is one of the smartest consumer AI integrations we have seen all year.

Every platform has copied Wrapped's static card format to death, leaving users numb to generic graphs. By turning raw behavioral data into a lively, narrative audio conversation, Spotify makes personal data feel conversational, human, and wildly shareable.

For Google, this is the ultimate Trojan horse. NotebookLM had earned a passionate cult following among researchers and students, but this partnership instantly exposes the Audio Overviews tech to millions of mainstream consumers who would never upload a research paper.

It sets a clear precedent for what end-of-year recaps should look like. Expect fitness trackers, banking apps, and ride-hailing services to copy this dynamic audio playbook next year.

The real test will be language support and accuracy. Right now, missing out on non-English music scenes and butchering indie artist names are obvious weak spots, but as a proof of concept, it is a masterclass in making generative AI actually fun.