Running AI in WebAssembly [Panel]

Oscar Spencer - F5 / Edoardo Vacchi - Red Hat / Bailey Hayes - Cosmonic / Ralph Squillace - Microsoft

AI is rapidly becoming part of every application stack. At the same time, WebAssembly is emerging as a universal runtime that runs anywhere, from the browser to the edge, servers, and even embedded devices.

But what happens when these two worlds meet?

In this panel, we’ll explore what it really means to run AI with WebAssembly. Can models run efficiently inside Wasm? What role do runtimes, GPU access, and WASI play? And how does Wasm compare with traditional approaches when deploying AI to the edge or inside applications?

We’ll discuss real-world use cases, current limitations, performance trade-offs, and where the ecosystem is heading next.