Image credits: spotify /
Spotify has already had success with its popular AI DJ feature, and now the music streaming service is bringing AI to playlist creation. The company on Monday introduced AI Playlists into beta, a new option that allows users to create a playlist based on written prompts.
The feature will initially be available to users on Android and iOS devices in the UK and Australia and will be rolled out in the coming months.
In addition to more standard playlist creation requests, such as those based on genre or time frame, Spotify's use of AI means people can request a variety of personalized playlists, such as “Songs to Sing to My Cat” or “Tunes to Fight the Apocalypse from.” “Zombies,” Spotify suggests. Prompts can reference all kinds of things, such as places, animals, activities, movie characters, colors, or emojis. The company points out that the best playlists are created with prompts that contain a range of genres and moods. Artists and contracts.
Spotify also leverages its understanding of users' tastes to customize the playlists it creates using this feature.
After creating a playlist, users can then use AI to review and improve the final result by issuing commands such as “less upbeat” or “more pop,” for example. Users can also swipe left on any song to remove it from the playlist.
In terms of technology, Spotify says it uses large language models (LLMs) to understand user intent. Spotify then uses its personalization technology — the information it has about the listener's history and preferences — to fulfill the demand and create a personalized, AI-generated playlist for the user.
The company uses a range of third-party tools for AI and machine learning experiments.
TechCrunch first reported in October 2023 that Spotify was developing AI playlists, when reverse engineers Chris Messina and Alessandro Paluzzi shared screenshots of code from the Spotify app that indicated AI playlists that were “based on your prompts.”
Spotify declined at the time to comment on the findings, saying it would not make a statement about potential new features. However, in December 2023, the company confirmed that it was testing AI-driven playlist creation after a TikTok video of the feature surfaced showing what a Spotify user described as “Spotify's ChatGPT”.
This feature can be found in the Your Library tab in the Spotify app by tapping the plus (+) button at the top right of the screen. A pop-up menu appears displaying Playlist AI as a new option alongside the existing Playlist and Mix options.
If a listener can't think of any prompts to try, Spotify offers quick suggestions to help people get started, like “Focus on work with electronic instruments,” “Fill the silence with café background music,” and “Enjoy the music.” Fun, upbeat, positive songs” or “Explore a niche genre like Witch House” and many more.
To save an AI playlist, tap the Create button to add it to the library.
The company points out that the AI has guardrails around it so it doesn't respond to offensive claims or those focused on current events or specific brands.
Spotify has been investing in artificial intelligence technology to improve its streaming service for months. With the launch of AI DJ, which expanded globally last year, the company used a combination of Sonantic and OpenAI technology to create an artificial version of the voice of Spotify's head of cultural partnerships, Xavier “X” Jernigan, who provides personalized song selections for Spotify. the user. Last year, Spotify said it was investing in internal research to better understand the latest AI models and big languages.
CEO Daniel Ek also explained to investors other ways Spotify can leverage AI, including summarizing podcasts, creating AI-generated audio ads, and more. The company also considered using artificial intelligence technology that would replicate a podcast host's voice for ads read by the host.
Before AI Playlists, Spotify launched a similar feature, Niche Mixes, that allowed users to create personalized playlists using prompts, but the product did not take advantage of AI technology and was more limited in terms of language understanding.