Foundations

Why Irys Exists

Before Irys, the team built and operated Bundlr, a data infrastructure layer used by hundreds of projects. That experience exposed recurring issues: unpredictable storage costs, the need to combine multiple storage and execution systems, fragile integrations between them, and limited guarantees around long-term data access and reuse.

None of these issues stood alone as a single failure. They appeared differently depending on the application, the scale, and the context.

At the same time, data usage patterns were changing in a way that affected nearly every sector of software.

By 2024, it was clear that data was no longer something applications generated occasionally and processed in isolation. Systems were beginning to produce data continuously, accumulate it over long periods, and depend on it to make future decisions.

AI made this visible, but it did not create the trend. It revealed it.

This shift exposed an assumption embedded in most blockchains at the time: that data would remain secondary to execution. Irys was designed with the opposite premise — that data would become an increasingly valuable and necessary resource in modern systems.

Irys was not built for a single application category or a specific failure case. It was designed as a base layer that removes assumptions about how data must behave. By treating data as a primary resource at the protocol level, Irys expands the design space available at the application layer. New systems can be built without inheriting constraints about where data must live, how it must be accessed, or how long it can persist.

If new problems emerge that the protocol does not solve directly, they can be implemented at the application layer while inheriting the base layer's storage, verification, and consensus guarantees.