The AI Era: The Urgent Economics of AI Music

Influence Over Consumption: Pioneering a New Economic Model for AI-Generated Music

AI-generated music is projected to generate a staggering $42 billion by 2028, according to CISAC’s latest study. With the rapid acceleration of AI platforms transforming the creative landscape, the music industry faces a critical moment to establish fair, scalable economic frameworks. Without proactive design, rightsholders risk being left behind in an era where technology outpaces traditional licensing structures.

In “Part 1” of this series, we introduced the foundational concept of sustainable business models for the AI era. Now, in “Part 2”, we take a closer look at the economic mechanics underpinning the Attribution Share business model. This framework offers the music industry a viable way forward, balancing innovation with equitable compensation. By introducing tiered attribution share models and leveraging familiar revenue structures, the model ensures rightsholders are positioned to thrive alongside AI platforms in this rapidly evolving landscape.

At the heart of this economic proposal lies an unprecedented reset opportunity for the industry—one that can finally rebalance the split of remuneration between recorded and publishing rightsholders. Unlike traditional streaming, where consumption volume dictated revenue allocation, AI-generated music shifts the focus to influence. The Attribution Share model recognizes the value of those who shape AI generative outputs, positioning publishers and songwriters to claim a greater share of the Industry Share (%). For publishers, this marks a chance to reverse the traditional market share revenue splits that record labels negotiated with platforms like Spotify and Apple Music during the early days of streaming.

This reset isn't just an evolution of the business model—it’s a transformation that redefines how value is attributed, aligning financial rewards with creative contribution.

Let’s dive into how Attribution Share works and why it holds the key to securing a fairer and more balanced future for the music industry in the AI era.

Attribution Share Calculation Models for AI Platforms

This comprehensive framework introduces tiered attribution share models that reflect familiar industry structures while accounting for the unique revenue potential of AI platforms. With real-time reporting, synthetic data protections, and premium catalog opportunities, it ensures equitable compensation for rights holders and dynamic growth for AI platforms.

NOTE: The provisional inputs have been purposefully oversimplified to illustrate the calculation mechanics.

The Mechanics

1. Transactional Model (Pay-per-Credit/Token):

Users purchase tokens to generate music, and the platform allocates a percentage (X%) of transaction revenues to rights holders. Attribution Share (Z$) is calculated based on the collective* influence (Y%) of a rightsholder’s assets.

Calculation Example:

• Per Unit Transaction Value: $0.10
• Number of Transactions (W#): 50,000,000
• Industry Share (X%): 60%
• Rightsholder A’s Influence (Y%): 30%
• Revenue Pool for Rights Holders:

• Attribution Share (Z$):

Proposed Formula:

2. Subscription Model (Recurring Fee):

The platform shares a percentage (X%) of total subscription revenues with rightsholders, distributed based on their collective* influence (Y%) within the generated outputs for the billing period.

Calculation Example:

• Monthly Subscription Fee per User: $10
• Total Subscription Revenue: $1,000,000
• Industry Share (X%): 65%
• Rightsholder A’s Influence (Y%): 25%
• Revenue Pool for Rights Holders:

• Attribution Share (Z$):

Proposed Formula:

3. Hybrid Model (Combination of Transactional + Subscription):

This model combines token-based and subscription revenues. Attribution Shares are calculated separately for each revenue stream before summing the totals.

Calculation Example:

• Total Subscription Revenue: $800,000
• Total Transactional Revenue: $200,000
• Industry Share (X%): 70%
• Rightsholder A’s Influence (Y%):
  • 20% on Subscription-based Outputs
  • 35% on Transactional Outputs

Attribution Share Calculation:

1. Subscription Pool:

2. Transactional Pool:

3. Total Attribution Share (Z$):

Sample Attribution Share Table & Definitions

*collective influence:

the compounded total of attributed influence, by "stacking" every individual output's share of influence that attributes to a given rightsholder (ie similar aggregation mechanic to user-centric streaming model's aggregation of each individual's listener share).

Alternative Calculation Provisions (Modeled After Streaming Language):

• Monthly Reporting Period:
  Attribution Shares (Z$) are calculated and paid monthly, ensuring rightsholders receive timely compensation based on real-time influence tracking.
• Influence Caps:
  • Economic: If a rightsholder’s influence (Y%) exceeds 50% of platform outputs during a reporting period, payout adjustments may apply to prevent market distortion, mirroring stream caps used by traditional streaming services.
  • Technical: Influence caps may be applied at the individual output generation level.
    • Example: for select assets from premium artist(s) or 'darling' catalog(s) – no AI platform output can be generated from select assets with an influence of 40% or higher due to 'substitutional' risk.
• Synthetic Data Safeguards:
  If synthetic datasets are used in AI model training, Attribution Shares will factor in estimated proportional influence, ensuring rightsholders remain compensated even when datasets extend beyond directly licensed material.
• Premium Tiering for High-Value Catalogs:
  AI platforms and rightsholders may introduce premium pricing for prompts that request specific high-value artists, catalogues, or repertoire. These tiered asset classes are defined through artist and rightsholder consent, allowing for differentiated pricing structures.
  Example:
  • Standard Prompt: Generates outputs based on non-premium catalog at regular token cost (e.g., $5/credit).
  • Premium Prompt: Requests specific artists or songs from a premium catalog, charging users at a higher token rate (e.g., $15/credit).
  • Attribution Share Impact:The premium prompt revenue is distributed with the same formula, but influence within the premium catalog is weighted more heavily. For instance, a 20% influence on a premium output could generate a higher payout compared to standard catalog outputs due to the higher per-credit price.
• Proposed Formula Adjustment for Premium Assets:

This ensures rights holders of high-value catalogs receive proportionate remuneration reflecting their asset class’s market significance, while platforms benefit from dynamic monetization opportunities.

Conclusion: A New Chapter in Music's Value Proposition

The advent of AI music generation marks a profound expansion in the value of music, creating opportunities to complement the advancements of streaming while unlocking entirely new commercial frontiers. 

By shifting the focus from consumption to influence, the Attribution Share model ensures that those who shape generative AI outputs are fairly and dynamically compensated. This framework doesn’t replace the economics of streaming; instead, it operates alongside it as a distinct model, addressing new opportunities and challenges presented by generative AI.

However, the stakes are high. The CISAC report warns that under current conditions, 24% of music creators’ revenues could be at risk by 2028, representing a cumulative loss of $10.5 billion over the next five years and an annual loss of $4.2 billion in 2028. These figures highlight the critical need for action: without the adoption of Attribution Share or similar frameworks, the rapid penetration of generative AI outputs could destabilize the revenue foundation that music creators rely on.

Together, Parts 1 and 2 of this series illustrate a vision for the future: preserving the progress made through streaming while addressing the existential risks posed by unchecked AI development. Attribution Share introduces scalable economic mechanics that align the music industry with the evolving technological landscape, ensuring that value expands proportionally with innovation.

This proposal is not a definitive solution—it is a foundational framework designed to empower stakeholders, economics experts, and licensing professionals to shape it further. It serves as an open blueprint, inviting the industry’s brightest minds to adapt it to their needs, building confidence in what can often feel like an uncertain path forward. The persistent question voiced by senior leaders—"But what is the business model?"—now has an answer, one grounded in fairness, scalability, and shared opportunity.

The projected $42 billion AI music market by 2028 is not just a number—it represents an unprecedented moment to redefine music’s worth in the digital age. Dr. Tamay Aykut of Sureel.ai puts it best, "the music industry needs accountability, not illusions." With the right structures in place, this evolution ensures that creators, publishers, labels, and platforms alike can thrive. The future of music is not confined to what has been; it is enriched by what can be.

Let’s transform uncertainty into opportunity and expand music’s value for generations to come.

[“Part 1” The AI Era: Building Sustainable AI Business Models for the Music Industry — originally published on October 29, 2024]