Suno CEO on AI Music, Creative Control and What Creators Should Watch

The most useful part of this interview was never a prediction about one version of Suno. It was the direction underneath it: AI music tools were moving from generation toward creator control.

That matters more now than it did when I first published this article. A better model can give you a better starting point. It still cannot decide what belongs in your song, what needs fixing, what should be released, or what rights and records you need to protect.

Start With the Source: Watch the Full Interview

Mikey Shulman joined Lightspeed partner Michael Mignano on the Generative Now podcast to discuss Suno’s growth, model quality, creator control, rights questions and possible directions for AI music.

Context matters: this interview reflects Suno’s position and ambitions at that point in 2025. When a founder talks about where a product is going, I treat that as useful signal—not a permanent product specification or guarantee.

1. Creative Control Was the Real Signal

The original article focused heavily on Suno 4.0, mobile creation and improved generation quality. Those details belong to a specific moment. The more durable idea was the push toward more control after and around generation.

That changes the creator’s role. If the only question is “Can the model make a song?”, then model quality dominates the conversation. Once creators can guide, compare, revise and edit, the better question becomes:

That is why my current process is INTENT → CREATE → SCORE → DECIDE → FIX → PACKAGE → PROTECT. The model participates in CREATE. The creator remains responsible for the rest.

2. Better Sound Is Useful. “Indistinguishable” Is Not a Useful Standard.

The 2025 version of this article said Suno 4.0 could produce tracks that sounded “almost indistinguishable from human-made songs.” That was too broad to present as a fact.

Model quality was improving quickly, and the interview reflected real optimism about that improvement. But creators do not need to win a human-versus-AI guessing game. They need a track that serves the project.

A polished output can still have the wrong hook, weak emotional movement, generic lyrics, an unsuitable vocal identity or a structure that never earns its payoff. Conversely, a rougher generation may contain the musical idea worth keeping.

3. Copyright Is Not a Feature Toggle

The original article also used an overly categorical sentence: “Fully AI-generated songs cannot currently be copyrighted unless there is significant human input.” That needs more precision.

Copyright eligibility depends on the human-authored material being claimed and the law of the relevant jurisdiction. Human-written lyrics, human composition, arrangement choices, performances, edits and other authored contributions can matter. A platform granting commercial-use rights is a separate question from whether copyright exists in every element of the finished track.

That separation is why I now use the labels AI-Generated, AI-Assisted and Full Copyright as practical creator categories rather than pretending every AI music project has the same rights profile.

4. Lower Friction Does Not Remove the Need for Skill

The original article said Suno was “democratizing music production.” I understand what I meant: tools like Suno can make it cheaper and faster for more people to hear an idea that might previously have stayed in their head.

But access is not the same thing as mastery, audience, ownership or a career.

Mobile creation and easier generation lower the friction of trying. They can help a songwriter test a direction, a non-instrumentalist communicate a concept, or an experienced producer compare alternatives faster. The creative value still comes from choosing what the project means and what survives the process.

5. Collaboration, Multimodal Creation and Blockchain Were Forward-Looking Ideas

The 2025 article listed live collaboration, richer multimodal inputs and blockchain-based tracking or royalty ideas alongside product developments. That presentation made speculation sound too much like a roadmap.

These ideas are worth preserving because they show how broad the conversation had already become. They should be read as possibilities discussed around the future of AI music, not as promises that every feature or business model would arrive exactly as imagined.

The blockchain point is a good example. Better provenance, attribution and royalty tracking are real creator problems. Blockchain is one possible technical approach; it is not itself a guaranteed revenue model.

What Aged Well—and What Needed an Update

The Creator Lesson I Would Keep

A platform’s roadmap is not your creative strategy.

New models, editing features and interfaces can expand what you are able to try. They do not answer the questions that determine whether the work becomes useful:

• What are you trying to make?
• What part of the output is actually worth keeping?
• What needs human revision or contribution?
• Is the track ready for the use you have in mind?
• What do you need to document before release?

That is where SCORE → DECIDE becomes more important as the models get better, not less important.

Where to Go Next

If you want current Suno guidance rather than a 2025 interview snapshot, use the Suno hub. If you want to compare Suno with the broader AI music workflow and decide which tool belongs where, start with the AI Music & Audio Hub.

Source credit: This article analyzes the Generative Now interview featuring Suno co-founder and CEO Mikey Shulman with Lightspeed partner Michael Mignano. The original interview remains embedded above so you can hear the discussion directly and form your own view.

Gary Whittaker
Founder and Operator, JackRighteous.com