Sources: OpenAI is developing AI tools to generate music from text and audio prompts, including capabilities such as adding guitar accompaniment to vocal tracks
OpenAI recently made a splash with artificial intelligence that generates short videos from text prompts—say, a scene from …
The Information
Context & Ripple Effects
OpenAI had already explored generative music with its Jukebox music model, while later coverage connects the company’s audio-model work to an audio-centric personal-device effort. The reported music tools extend that audio arc from generation toward modifying an existing vocal recording.
The development also places OpenAI alongside voice-and-audio specialists: ElevenLabs had previewed a model for text-prompted lyrics and song samples. What matters here is the reported combination of text prompting and audio-input editing, rather than music generation alone.
First-order effects
OpenAI’s audio-model roadmap would broaden to include music creation and accompaniment workflows, if the reported tools reach users.
Creators could potentially use a vocal track as input and request instrumental additions, reducing the need to start every musical element from scratch.
Second-order effects
Audio-AI rivals and music-creation software providers would face pressure to match prompt-based editing features, not just standalone synthetic voice or song generation.
The value of music tools would shift toward control over supplied audio—such as preserving or transforming a vocal track—where workflow fit can matter as much as raw generation quality.
Third-order effects
If audio models increasingly handle generation, editing, and interaction in one stack, music creation could become part of a broader multimodal assistant layer rather than a separate specialist category.
That convergence would make commercialization and rights-sensitive deployment central differentiators for AI music products, though the report does not establish how OpenAI would package or release these capabilities.
The trend:Generative AI is moving from producing isolated media outputs toward workflow-native tools that can take existing audio as context and alter it on request.
@nearcyan this one is particularly funny because i'm pretty sure google has had the best music generator for a while but just not released due to copyright concerns
Interesting and not surprising that OpenAI is working on a new music generating model. They were one of the first, with Jukebox. The Sora 2 model can generate music already within videos. Suno is generating over $100 million a year in revenue just for a music generator, making
OpenAI tunes up generative music (and rivals Suno/Udio) OpenAI is diving into AI-generated music, working with Juilliard students to train models that can turn text or audio prompts into full songs. The move follows its viral video tool Sora and signals a push to rival Google's […
@btibor91 The copyright problem here is massive. Music labels already sued Suno and Udio for training on copyrighted content. OpenAI needs licensing deals before launch or faces same lawsuits. Training data sourcing determines viability more than model quality.
OpenAI just dropped the bass. The company that changed how we write, code, and create images is now tuning its sights on music. OpenAI is reportedly building an AI music generator that could rival Suno and Udio. Think about that for a second. ChatGPT turned words into meaning. [i…
Suno is great. It can probably write every future Imagine Dragons or Taylor Swift song. But I imagine lawyers at every major record label are strategizing and salivating to nail their IP infractions. OpenAI's advantage in this space is the ability to hide its tracks on IP theft. …
Music industry will be massively disrupted by AI, no one knows what's coming. why? 1/ music is a huge market and running on obsolete tech 2/ music is a creative field and AI is super creative, match made in heaven but how can artists benefit? sell AI-use licenses via
According to The Information, OpenAI is developing AI that generates music and has been working with Juilliard School students to annotate music scores, based on reports from two people familiar with the project The company has discussed building tools that let users generate [im…
OAI never released any of their past music models past the original juekboxes which were alright, but now I suppose they will want a new one which is not just fully neural e2e text->audio but allows you to have a suno-like experience. makes sense https://x.com/...
I am one of them; $100 of their annual revenue comes from me ☺️ @SunoMusic v5.0 is great, and I mostly listen to AI-generated music created there my own and by others as well.
- Background music for youtubers / short form creators - Ads with music with funny lyrics - Production Studios (I know people at huge prod studios who are secretly obsessed with AI generated music) - Mass Uploading to Spotify and gaming the payout system (Slowly becoming harder,
@unusual_whales AI writing music isn't disruption, it's vertical integration of creativity. When models compose culture, the moat shifts from talent to distribution. Who controls the soundtrack when algorithms own the studio?
Open AI is training a music model. As far as I know this would be the first since Jukebox in 2020. For a long time it looked like they wanted to completely avoid the huge legal fight with the labels, but something must have changed their mind. This model will need a good name.
My 74 year old dad writes original lyrics and pays for a Suno subscription to make songs for my 97 year old grandma and also friends in his congregation. He's just the most unique case I know of, there are dozens others I know about including some of the most famous household
Kinda seems like OpenAI is just throwing out every hideous idea they can think of in a quest to find the one that will bring in a trillion dollars, the inevitable end point of which is them giving up and saying “OK, porn it is. We're doing AI porn.” [embedded post]