Artificial intelligence continues to reshape the music industry. In late 2025, Warner Music reached a licensing agreement with AI music company Suno after settling a lawsuit, a move that signaled a new phase in how major labels approach generative tools.
Some view the deal as progress. Others see it as a consolidation of power between large rights holders and fast-growing AI firms.
At the same time, blockchain-based music platforms claim they can solve attribution, consent, and compensation challenges that AI systems have intensified.
Dzmitry Saksonau, founder of the AI-native social music platform JGGL, argues that licensing agreements alone do not resolve structural imbalances in the industry.
In this interview with CCN, he discusses the limits of AI provenance, the role of Ethereum, token incentives, and what failure could look like for AI-driven music platforms.
Suno-Warner Deal: Progress or Reinforcement of Old Power Structures?
Warner Music’s agreement with Suno shows that major labels are willing to license AI companies rather than fight them in court.
Some argue that this signals institutional acceptance of AI in music production. Others question whether it strengthens incumbents.
Saksonau sees both sides.
“It’s progress in the sense that the industry is finally acknowledging AI as a permanent part of music production,” he says.
“But a few major rights holders still run the negotiations. Everyone else gets whatever terms they agree on. Independent creators and smaller contributors don’t sit at that table.”
For him, the issue is not whether licensing deals should exist. It is about who controls the terms.
“Web3’s promise isn’t to block licensing deals – it’s to make attribution, consent, and compensation more granular and creator-driven,” he says.
“The risk is that without new infrastructure, AI simply becomes another layer controlled by incumbents rather than a tool that meaningfully redistributes power.”
In other words, AI can either widen participation or reinforce centralization, depending on the systems built around it.
AI Provenance and the Limits of Traceability
One of the most complex debates around generative AI concerns provenance. Provenance refers to the origin and history of a creative work, including who contributed and under what terms.
Most AI music models train on vast datasets. Their outputs are probabilistic, meaning they generate results based on patterns rather than copying a specific input.
As a result, it is nearly impossible to trace a particular output back to a specific training file.
Saksonau believes the industry often frames the problem incorrectly.
“Provenance in AI music shouldn’t be framed as ‘this output equals that input.’ That’s not how generative systems work,” he says.
“Instead, the focus should be on documenting process, consent, and contribution at the creation stage.”
JGGL approaches provenance at the moment of collaboration and publication, not at the model-training stage.
“JGGL records provenance at the point of action: when someone contributes, collaborates, or publishes,” he explains.
“That record captures who was involved, what terms applied, and where revenue goes. It doesn’t attempt to trace every training-data input back to a specific output, which no system can do reliably.”
This distinction matters. Rather than solving the unsolvable question of model memory, JGGL focuses on verifiable authorship and contribution once a track exists.
Who Is Responsible for AI Training Practices?
Blockchain can record ownership and contribution splits after a track is created. It cannot verify whether an AI model was trained on copyrighted material without the copyright holder’s consent.
That gap raises a question of responsibility.
“Responsibility is shared,” Saksonau says.
“Model developers are responsible for how systems are trained and what claims they make about compliance. Platforms are responsible for how those tools are integrated and disclosed. Creators are responsible for how they choose to use them.”
He makes a clear distinction between upstream and downstream accountability.
“Blockchain doesn’t solve upstream training issues, but it does make downstream behavior auditable,” he says.
“That distinction matters – especially as regulation increasingly focuses on accountability rather than technical purity.”
In practical terms, this means that while blockchain cannot fix past data ingestion practices, it can make present and future use more transparent.
Oversupply, AI, and the Value of Human Creativity
Many artists fear that AI-generated music will flood the market, reducing the value of human work.
Monetization rails, meaning built-in systems that allow creators to earn from their content, could accelerate production even further.
Saksonau argues that oversupply already exists.
“Oversupply already exists; AI just makes it more visible,” he says. “The real question is whether systems can differentiate contribution rather than volume.”
He believes the real problem lies in opaque systems that fail to recognize creative judgment.
“Monetization rails don’t inherently devalue creativity – opaque systems do,” he says.
“When the record shows who chose the references, who adjusted the prompt, and who edited the final piece, each of those people can get paid.”
He points to a gap in many AI tools.
“Right now, most tools only track who clicked ‘generate.’ That’s the gap. Making contributions visible is what lets you pay for judgment, not just computation.”
In his view, transparency can shift value from raw output to human input, including curation, editing, and creative direction.
Why Web3 Music Platforms Struggled During the NFT Boom
During the non-fungible token (NFT) boom, many Web3 music platforms promised fairer economics for artists. Most struggled with adoption once speculative interest cooled.
Saksonau attributes the failure to multiple factors.
“All of the above, but structurally, the biggest issue was abstraction,” he says.
“Most platforms asked artists to care about tokens and wallets before solving everyday creative problems.”
NFT stands for non-fungible token, a unique blockchain-based digital asset that represents ownership of a specific item, such as music or art. While NFTs introduced new monetization models, many artists found the technical barriers difficult.
“Fair economics only matter if creators can actually make, collaborate, and release music more easily,” Saksonau says. “Speculation amplified that disconnect.”
He believes future success depends on invisibility.
“If Web3 succeeds in music, it will be invisible – embedded into workflows rather than marketed as an ideology.”
In other words, creators should not need to understand blockchain mechanics to benefit from them.
The “Open Studio” Model and Power Imbalances
JGGL describes itself as an “open studio.” In collaborative environments, contributors often have unequal bargaining power, visibility, or industry connections.
Saksonau argues that clarity at the start reduces exploitation.
“By making terms explicit by default,” he says.
“Every contribution on JGGL carries defined attribution and splits from the outset, rather than relying on informal agreements or post-hoc negotiations.”
Contribution splits refer to the percentage of revenue each participant receives from a track.
“Open systems still have rules,” he says. “The difference is that those rules, and the decisions behind them, are visible to participants.”
Transparency does not eliminate inequality, but it can limit hidden arrangements.
“When contributors can see how value flows before they participate, power imbalances are reduced, even if they don’t disappear entirely.”
The emphasis remains on visibility rather than trust alone.
Tokens and the Risk of Speculation
Many blockchain platforms introduce native tokens, which can be traded or used within an ecosystem. Critics argue that tokens can distort creative spaces by encouraging speculation over artistry.
JGGL uses its native token, $JGGL, to power transactions and governance.
“The token gives members access and voting rights. The actual product is the protocol itself,” Saksonau says. “If speculation becomes the main reason people engage, the platform has failed.”
He frames tokens as infrastructure rather than entertainment.
“Our design assumption is that most users care about creation, collaboration, and recognition first,” he says.
“Tokens exist to support transactions – splits, tipping, access – not to gamify culture.”
He adds a warning.
“The moment financial abstraction overtakes creative intent, you lose the community you’re trying to serve.”
In this model, the token underpins attribution and revenue flows rather than functioning as a speculative asset.
Why Build on Ethereum Despite Costs and Scalability Issues?
Ethereum, a decentralized blockchain network, has faced criticism over scalability constraints. Saksonau acknowledges there are trade-offs but emphasizes credibility.
“Ethereum isn’t perfect, but it has the strongest guarantees around neutrality, composability, and long-term credibility,” he says.
Composability means that applications built on Ethereum can interact with each other seamlessly.
“For ownership and attribution, those properties matter more than raw speed,” he says.
“We’re pragmatic about scaling – using layers and abstractions where appropriate – but anchoring provenance to a widely trusted settlement layer reduces fragmentation and platform risk over time.”
A settlement layer refers to the base blockchain that finalizes transactions permanently.
In his view, trust and durability outweigh performance gains from purely off-chain systems.
Disputes, Arbitration, and the Role of Law
Smart contracts, which are self-executing programs on a blockchain, can automate revenue splits. However, disputes over authorship or intent can still arise.
Saksonau does not argue that code replaces courts.
“Code sets the defaults,” he says. “If two collaborators disagree over revenue splits, the on-chain record shows what each person contributed and what terms they accepted.”
That record can streamline resolution.
“That record makes arbitration faster, whether it happens through the platform or a court,” he says. “It doesn’t remove the need for legal systems; it gives them clearer evidence to work with.”
He contrasts this with traditional workflows.
“The difference is that evidence is explicit and shared, not buried in emails or informal chats.”
Blockchain, in this framing, serves as structured documentation rather than a replacement for legal institutions.
What Failure Would Look Like for AI-Native Music Platforms
As AI-native platforms expand, assumptions about creator behavior may prove wrong. Saksonau reflects on what failure might look like within five years.
“Failure would mean creators revert to closed platforms because open systems are too complex or extractive,” he says.
He challenges a common belief in crypto circles.
“The most common wrong assumption is that creators want to be investors first and artists second,” he says. “Most don’t. They want tools that respect their work, save time, and pay fairly.”
If platforms prioritize financial novelty over usability, he expects a quiet decline.
“If AI-native platforms forget that – and optimize for financial novelty over creative utility – they’ll disappear quietly, not dramatically.”
The warning is clear: technology must serve creative workflows, not the other way around.
As AI tools become standard in music production and licensing deals reshape industry norms, debates around provenance, compensation, and power continue to intensify.
For Saksonau, the solution lies less in ideological positioning and more in infrastructure that documents contributions clearly and distributes value transparently.
Whether blockchain-based systems such as JGGL can move beyond speculation cycles and embed themselves into everyday creative practice remains an open question.
What seems certain is that AI will not exit the music industry. The question is who sets the rules, and how visible those rules become.
Read more on CCN – Capital & Celeb News
