Spotify teams up with Sony, UMG, WMG, Merlin and Believe on artist-first AI music products

Spotify teams up with Sony, UMG, WMG, Merlin, and Believe to build "artist-first" AI music products

On October 16, 2025, Spotify announced new partnerships with Sony Music Group, Universal Music Group, Warner Music Group, Merlin, and Believe to develop AI music products under a clear promise: put artists and songwriters first.

The company is building a generative AI research lab and product team, and says new offerings will be created under upfront agreements with rightsholders. This follows new AI policies and a crackdown that removed 75 million spam tracks over the past year.

Why this matters for product teams

It sets a template for shipping AI features with consent, licensing, and compensation baked in.
It signals a shift from "move fast and ask forgiveness" to "secure rights first, launch second."
It elevates creator controls and attribution as core product requirements, not add-ons.

The four principles, translated into execution

Upfront partnerships: Build new AI features only after direct licensing with labels, publishers, and distributors. Product implication: legal and product must co-own the launch plan.
Choice in participation: Artists and rightsholders choose if and how they participate. Product implication: offer opt-in, usage scopes, and clear controls per feature.
Fair compensation and new revenue: AI features should create new income streams with transparent credit. Product implication: design for attribution, usage metering, and payout logic from day one.
Artist-fan connection: AI augments creativity; it doesn't replace it. Product implication: focus on co-creation, discovery, and engagement features that spotlight human creators.

What Spotify is building

A gen AI research lab and product team combining internal R&D with external tech partnerships.
Creator-informed development, with artists, producers, and songwriters consulted on tools and policies.
Guardrails: content integrity, provenance, and enforcement consistent with its recent policy moves.

Where the industry stands (in their own words)

Spotify's Alex Norström: "Technology should always serve artists, not the other way around."
Spotify's Gustav Söderström: AI is a major shift; Spotify plans to build it with the industry and grow its AI team.
Sony's Rob Stringer: direct licensing before launch is "the only appropriate way" to build these products.
UMG's Sir Lucian Grainge: Gen AI must sit inside a commercial model where all parties can flourish.
WMG's Robert Kyncl: new AI licensing deals should protect and compensate rightsholders.
Believe's Denis Ladegaillerie: prioritize "responsible AI" and "value-creative AI" that fuels artist development.
Merlin's Charlie Lexton: partner with companies that respect copyright and enrich, not displace, artists.

What this means for your AI product roadmap

Data and rights strategy: Treat datasets, training rights, and output licensing as first-class requirements. Build a licensing intake pipeline and contract-aware data catalogs.
Participation controls: Offer granular opt-in, content scope (training vs. style transfer vs. remix), and revocation. Maintain audit trails.
Attribution and payouts: Track contributions across datasets, prompts, stems, and co-creations. Expose credits in UI; meter usage for payouts.
Abuse prevention: Integrate content authenticity checks, spam detection, provenance metadata, and model use policies.
Co-creation features: Build tools that help creators iterate faster, test concepts, and engage fans without eclipsing human authorship.

Metrics to instrument from day one

Creator opt-in rate, feature-level participation, and retention.
Share of plays or sessions influenced by AI features.
Average new revenue per participating artist/songwriter.
Attribution completeness and dispute resolution time.
Policy violations per 1,000 creations and time-to-remediation.
Licensing cycle time from deal approval to product availability.

Risk and governance checklist

Copyright-safe training and output generation; document provenance and consent.
Clear style cloning boundaries and name/likeness protections.
Transparent user labeling for AI-assisted content.
Cross-functional review across product, legal, trust & safety, and comms.

For context on policy directions, see the Spotify Newsroom and the U.S. Copyright Office's AI resources here.

What to watch next

Spotify's first "artist-first" AI launches: creator controls, attribution UX, and payout models.
New licensing frameworks for training, style simulation, and co-creation.
Tooling for provenance and spam reduction at catalog scale.
Expansion to more rightsholders and potential APIs for partners.

Action steps for product leaders

Stand up a rights and data intake pipeline with model cards tied to licensing terms.
Ship creator participation controls before public rollout; default to consent.
Design attribution and metering early-retrofitting later is costly.
Define abuse policies with automated detection and human review loops.
Run creator councils for feedback on incentives, UX, and policy clarity.
Upskill teams on AI product, policy, and safety. Explore role-based learning here.

The signal is clear: AI music products will be built with consent, credit, and compensation at the core. Teams that operationalize these principles now will ship faster with fewer surprises later.