Chemical engineering has long been about optimization, scalability, and sustainability. By embedding those same principles into the architecture of SDLs, we can build a future of accelerated, reproducible, and democratized discovery.

This strategy aligns resource efficiency with discovery speed and empowers AI agents with richer, more reliable data streams.

Rather than retrofitting conventional reactors and instruments into autonomous workflows, we advocate for a bottom-up SDL design that treats every unit operation not just as a chemical transformation step, but also as a data-generating process module.

In this article, we argue that engineering principles, especially Process Intensification (PI), must serve as the foundation for SDL design. Why? Because PI enables '𝐃𝐚𝐭𝐚 𝐈𝐧𝐭𝐞𝐧𝐬𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧', that is the key to unlocking the full decision-making power of AI agents in SDLs.

As self-driving laboratories (SDLs) continue to gain momentum in chemistry and materials research, it’s critical to ask: Are we just automating yesterday’s tools, or are we rethinking experimentation from the ground up?