Dear reader,

When Universal Music Group filed a lawsuit against AI music generators over the summer, a critical question emerged: who owns a musician's data? It’s a question with far-reaching implications, not only for record labels but also for any business that relies on third-party distribution channels or supply chains. Data control will determine the future leaders of entire industries. And the music industry’s data ownership questions lay bare the profound consequences every industry faces (or will soon face).

Historically, major labels like Universal Music Group (UMG) served as the bridge between artists and their audiences. Now, streaming platforms (like Spotify, Apple Music, and Meta) have taken over that role—collecting an incredible amount of data, including listener demographics, behavioral patterns, and social graphs, giving them unmatched insights into consumer preferences and the future of the music industry. 

This is a massive shift. If platforms understand audience preferences better than music labels, and they control the relationship with the consumer, where does the label fit in the value chain? Moreover, if streaming platforms can leverage data to generate highly personalized and individualized synthetic music (using genAI) based on their tastes, will this be “good enough” for social media and diminish the demand for artists? The outcome is a growing power imbalance, where streaming services can leverage their data to disrupt the role of major record labels, and potentially replace the labels altogether.

For most industries, relevance will require retaining control, if not outright ownership, of data. At Opaque, we believe that confidential AI can play a pivotal role in restoring data sovereignty to all. 

In this issue, we’ll dive into how confidential AI can enable data sovereignty in the music industry and explore the broader implications for other industries facing similar challenges. It’s a story of turning a one-way data flow into a mutually beneficial partnership.

I hope you find this perspective insightful, and I’m eager to hear your thoughts on where this journey might lead.

— Aaron Fulkerson, CEO, Opaque Systems

Restoring Data Sovereignty: How Confidential AI Could Transform the Music Industry—and Beyond

Data has become one of the most valuable assets a business can possess. Knowing how to protect and leverage it is essential. This is particularly true for industries like music, where data about audiences, listening habits, and engagement holds the key to shaping the next big hit. But as music distribution has moved to streaming platforms like Spotify, Apple Music, and Meta, traditional record labels face a new challenge: retaining control over the data that defines their business.

For decades, music labels have played a central role in the industry, offering artists the support they need to reach audiences. Today, that support comes with a caveat. As music is distributed through streaming services, those platforms gain access to extensive data about listeners—everything from what they listen to, how long they engage, and what they’re likely to play next. This data provides streaming platforms with insights that go beyond what the labels themselves can access, creating a shift in the power dynamic.

The implications of this shift are profound. Streaming platforms can use the data they collect to train AI models that replicate popular artists' styles, potentially creating music that directly competes with the very artists the labels represent. For labels, this poses an existential threat: if platforms control the data and can leverage it for creative purposes, the role of labels in the industry could diminish significantly. But a powerful solution exists—one that lies in maintaining control of data through confidential AI.

Confidential AI: A Path to Data Sovereignty 

Confidential AI offers a way for music labels to reclaim their data sovereignty without sacrificing the benefits of digital distribution. It allows labels to share data with streaming platforms in a secure manner, ensuring that while the data can be used for necessary analysis and personalization, the platforms cannot access the underlying information itself. This means labels retain control over their data while still taking advantage of the reach that streaming services provide.

With confidential AI, the risk of genAI models mimicking top artists is significantly reduced. Labels can continue to use data to understand audience preferences and shape new releases without fear of inadvertently training a competitor. It turns data protection into a strategic asset, allowing labels to stay competitive and innovative in a market increasingly shaped by AI.

Beyond Music: Data Sovereignty for High Tech and Insurance 

The need for data sovereignty extends well beyond the music industry. In the high-tech sector, data is the fuel that powers everything from product development to customer engagement. Yet, sharing this data with external partners for analysis or integration can be fraught with risks. Confidential AI provides a way to collaborate on data without exposing it, allowing high-tech companies to unlock new insights while keeping control over their proprietary information.

For example, a tech company could partner with an AI firm to analyze user behavior and improve a product’s performance. With confidential AI, they can ensure that the AI firm processes the data without gaining access to the sensitive details. This fosters innovation while maintaining the company’s competitive edge, creating a win-win scenario for both parties.

In the insurance industry, data sovereignty is even more critical. Insurers handle vast amounts of sensitive data, from medical records to personal information, to assess risk and tailor policies. Sharing this data with partners for analytics or fraud detection has traditionally required a delicate balance between security and transparency. Confidential AI can ensure that insurers get the insights they need without compromising data privacy, enabling them to stay compliant with regulations while delivering advanced, personalized products to their clients.

A Strategic Imperative for the Future 

As AI and data-driven insights become more integral to business strategy, the ability to maintain data sovereignty is no longer optional—it’s a strategic imperative. Confidential AI allows companies to turn data security into a source of competitive advantage, providing the flexibility to innovate without fear of data misuse.

For the music industry, it means labels can continue to support their artists while retaining control over the insights that drive creative decisions. In high-tech and insurance, it means businesses can explore new opportunities for AI-driven growth without sacrificing data control. Across industries, confidential AI is enabling companies to transform data protection from a defensive posture into a pathway for sustainable success.

In the Lab

The Latest Happenings at Opaque Systems

NVIDIA AI Summit: Opaque Concludes Two Days of Networking 
Last week, a team from Opaque—Aaron Fulkerson, Jason Lazarski, Jamie Aliperti, David Thornbury, and Keith Haller—wrapped up two productive days at the NVIDIA AI Summit. They connected with attendees to showcase how Opaque's platform accelerates AI deployment by unlocking the value of sensitive data while ensuring verifiable privacy and sovereignty. The team had engaging and insightful conversations, generating significant excitement around Opaque’s confidential AI solution.

NVIDIA AI Summit: Opaque’s Confidential GPU Demo
Opaque's confidential AI platform enables the training of models and running of inference at scale using confidential GPUs, which accelerate AI workloads on sensitive data while ensuring verifiable privacy and data sovereignty. Recently demoed at the NVIDIA AI Summit, Opaque’s technology empowers industries such as high tech, finance, and insurance to securely harness their most valuable data for AI-driven innovations. This demonstration highlighted the platform's potential to transform how organizations manage and utilize sensitive information.

Feature: Aaron Fulkerson on What Lies Ahead for AI in Cybersecurity
AI is gaining recognition for improving cybersecurity and addressing the skills gap across industries, but the rise of GenAI is prompting concerns about the limits of traditional privacy practices, calling for a fresh approach. In a roundup article, Help Net Security featured Aaron Fulkerson’s discussion on how the weaponization of GenAI has made existing data privacy practices—like masking, anonymization, and tokenization—obsolete.

Code for Thought

Worthwhile Reads

💸 Business tech leaders are looking to put their AI dollars toward results. Many companies that experimented with AI are now facing pressure to show real returns, with around 90% of generative AI projects struggling to move beyond the lab due to issues with accuracy and reliability. A key challenge is ensuring broader access to corporate data so AI models can perform effectively in real-world applications. By enabling sovereign access to sensitive data, confidential AI solutions can help businesses overcome these hurdles and unlock the full potential of their AI initiatives.

☁️ It’s boom time for the cloud. Along with its sister report, the Thales Global Data Threat Report, the 2024 Thales Cloud Security Study looks into aspects of cloud security and revisits the impact of a dynamically expanding and complex attack surface. Of note, 47% of data stored in the cloud is sensitive, yet less than 10% of enterprises have encrypted 80% or more of their cloud data. Unprecedented demand for compute and a growing volume of data challenge priorities for achieving better, more secure cloud adoption. 

💡 Utilities are looking to a new generation of technologies. AI is helping utility companies predict outages and manage infrastructure more efficiently, improving public safety and reliability during emergencies. With more diverse data sources, utilities can leverage historical data trends to optimize grid performance and prevent outages. However—particularly in emergency situations—there’s a need for transparency and ethics during AI deployment, and utility companies require certain tools and frameworks to make fair and unbiased decisions. Confidential AI solutions enable access to sensitive and diverse data sources, ensuring transparency and ethical decision-making while supporting a stronger, more resilient grid. 

💼 The role of corporate governance in the AI discussion. A new report released this week from the National Association of Corporate Directors marks the first formal acknowledgment of the critical role of corporate governance in emerging technologies like AI, concluding that effective corporate governance will be key to whether new technologies—such as AI—gain acceptance and deliver value for organizations, economies, and societies, especially in regulated industries like financial and insurance services. As corporate governance sharpens its focus on AI oversight, the importance of adopting ethical, scalable, and responsible AI systems becomes clear. Businesses can meet higher standards for governance while balancing innovation with the need to protect sensitive data, with solutions like Opaque.

🔢 Countering monopolies in the AI search market. The U.S. Department of Justice has proposed significant antitrust measures against Google to dismantle its search monopoly, potentially impacting its revenue and stalling advancements in AI. These proposals include divesting key assets like Chrome and Android, as well as imposing restrictions on data collection that could allow competitors more room to grow. As Google faces increased pressure from rising AI competitors like OpenAI and Microsoft, utilizing solutions like confidential AI can help businesses navigate regulatory challenges while protecting user data, ultimately fostering innovation in a compliant manner.