Fifteen thousand years ago, an emptied African lake refilled, and one fish species exploded into five hundred. Something similarly radical is happening to data center infrastructure, and most of the industry is still pricing it with a real estate calculator. AI factories don't reward the best single design. They reward the operator who can prove, with evidence instead of a spec sheet, that the next site will perform exactly like the last one did.

AI workloads are reshaping grid dynamics: massive data centers can drop or swing hundreds of megawatts in seconds, threatening the stability of the power network. To address this, developers are exploring foundational changes to data center design and reinventing the way they are delivered.

Recent cancellations of major data center projects don’t repudiate the boom. They reveal a need for a new approach.

Can small-scale data center deployments help commercial property owners with persistent vacancy gaps? Perhaps, if we can reframe the problem from one of filling space to one of maximizing assets.

The decline of white space is upon us. As AI demands skyrocket, conventional data centers face their Kodak moment. In order to thrive, companies must adapt to a server rack that individually requires building-scale power and cooling and drives the design of the facility around it.

The number of AI use cases in design is growing. To evaluate them, we need to understand the technology’s strengths and limitations. To track them, we can map them against our project workflows.

Reflections from Las Vegas. The BIM revolution offers us valuable lessons. Designers and consultants should wield them as we navigate the AI revolution.

Innovation doesn’t always involve invention. Many times, it’s more about establishing a deep connection to what’s really needed.

History shows that structural forms weren't arbitrarily used—they emerged in response to specific challenges and constraints. As designers, we should strive to do the same in our work today.