The IRIS project — Interface Reasoning for Interactive Systems — will provide a systematic, rigorous theory for reasoning about how complex, large-scale systems communicate with one another and do so in a way which supports the development and deployment of verification tools. IRIS builds upon recent spectacular successes, originated and led by its team, in program verification. These now enable us for the first time to address issues that present at the level of the large, complex, dynamic, highly-distributed systems that we now critically depend upon in our daily and commercial lives. IRIS will develop the techniques pioneered for programs so that they can be applied in much wider scope to deliver much more robust systems ecosystems. This will be achieved through interconnected and mutually supporting component projects, drawing upon common foundations, that span theory and application, across the range of system scales: systems security; general code interfacing; reasoning about protocols; and policy and organizational management.