Katy Perry’s recent pushback against the White House for using her hit song “Firework” without permission underscores a friction point that will only grow sharper as digital content proliferates and AI tools make it easier to remix, repurpose, and deploy media at scale. The fact that Perry had just sold the song as part of a $225 million catalog deal makes the incident even more pointed. Someone owns the rights. Someone should have asked first.
This is not a new problem, but it is becoming a more complex one. When a single piece of music can be licensed to different parties, embedded in different projects, and claimed by different stakeholders, the administrative surface grows. Add AI into the mix, systems that can generate, modify, or contextualize content automatically, and the tracking and enforcement of rights becomes less a matter of goodwill and more a matter of infrastructure.
Behind every digital asset is a chain of custody. Who owns it? Who can use it? Under what terms? When does a license expire? These questions used to be managed through contracts and manual oversight. Today, they demand smarter systems. Rights management platforms, metadata standards, API integrations that connect licensing data to production workflows, and audit trails that log every use, these are not nice-to-haves anymore. They are business essentials.
The same principle applies wherever digital property changes hands at scale. A $225 million catalog deal is not just a financial transaction; it is a transfer of thousands of permissions and restrictions that now have to be tracked, enforced, and sometimes defended. That requires software built with the same rigor you would expect from a bank or a healthcare provider: architecture that scales, security that holds, integration that does not break downstream workflows, and logging that survives an audit.
AI makes this challenge both more urgent and more solvable. Intelligent systems can monitor where content is used, flag unauthorized deployments in real time, and help rights holders understand their own assets more clearly. But those systems themselves must be engineered as production software, not as one-off demos or quick proofs of concept. They have to be secure, maintainable, and reliable enough to protect assets worth hundreds of millions of dollars.
If you are thinking about building smarter systems to track, manage, or enforce rights across digital content, or if you have any custom software or AI challenge that has to survive real-world complexity, that is exactly the kind of problem where production-grade engineering matters most. Start a conversation with ABIE. Email [email protected] and tell us what you are trying to build.