DriveNets Turns Two Data Centers 52 Miles Apart Into One GPU Supercluster
DriveNets is running two GPU data centers 52 miles apart as a single training cluster, arguably the clearest technical answer yet to the community-opposition wave blocking AI datacenter buildouts. Also today: Cisco and Extreme both bet the network itself is where AI agents should hold real operational authority, without answering who's accountable when they're wrong.
Welcome to Amaze Networks for Monday, July thirteenth. Quick question to start the week: if community opposition can now block or delay one hundred thirty billion dollars in data center projects in a single quarter, is the industry's real answer to fight harder for land and power, or is it something else entirely? We've got a story today that argues pretty hard for something else entirely.
I think I know exactly where this is going, and I like it. Friday we said opposition had become a bigger capex-blocking force this year than power constraints or permitting delays individually.
We did say that. And today there's a company that seems to have taken that as a design brief instead of a warning.
New episodes, every weekday.
Amaze Networks drops at 4 AM CT, Monday through Friday. Spotify and Apple Podcasts submissions in progress.