Google Chief Technologist for AI Infrastructure recently said that the company’s main advantage in the AI race lies in how it builds and runs technology across the entire stack, from chips to software. Speaking at the Cisco AI Summit in San Francisco, Amin Vahdat said Google’s ability to co-design hardware, models, and infrastructure internally allows it to move faster than rivals. “The real secret weapon that we have is that we get to work together across the stack at the company to solve the end problem,” Vahdat explained, adding “If you look at TPUs, they’re not designed in isolation. They’re co-designed with DeepMind, but also taking input from all of the different use cases: search, ads, YouTube, Cloud.”His remarks come as competition intensifies with companies such as OpenAI and Nvidia, and as AI capabilities improve at a pace that Vahdat said now exceeds traditional computing trends.“Things feel like they’re getting twice as good, even faster than Moore’s Law,” Vahdat said. “At this point, I don’t see any slowdown,” he stated.

Google plans data centers in space

During the summit, Vahdat also discussed Google’s moonshot to build AI data centers in orbit. Announced in November 2025, Project Suncatcher is aimed at harnessing sun’s energy directly in space. “From a first principles perspective, it holds a lot of appeal,” Vahdat said. “A sun-synchronous orbit with 24/7 solar power—no batteries needed, no cloud cover, no sunset. Networking in space gets you a 50% reduction in latency for speed of light, not having to go through fibers.”Vahdat’s comments followed the launch of Google’s Gemini 3 model, which posted strong results across several AI benchmarks. He said the model’s performance supports Google’s long-term strategy of investing in its own models, chips, and data centres. “Gemini 3 has been awesome. We’ve been on this three-plus year journey at Google. I’m always rooting for the underdog, and it was great to be at a place where Google was the underdog,” he said on the new AI model.