🤖 AI Summary
Cerebras CEO Andrew Feldman warned that the U.S.’s decentralized policymaking is slowing AI infrastructure buildout and ceding ground to China, which he says has used centralized decision‑making to rapidly scale energy projects—dams, coal plants and massive solar—to power large AI deployments. Feldman argues that a “patchwork” of local ordinances and permitting can stall projects and cost “billions,” and he proposed a federal fix: a 5‑year moratorium on state and local AI regulations (with a renewal option) to prevent each company from having to navigate different rules in every jurisdiction—a “tax on innovation” that slows data center and power capacity expansion critical to training and running large models.
The comments matter because they connect policy, energy infrastructure and semiconductor strategy to competitive AI advantage. Cerebras, which makes purpose‑built, plate‑sized AI chips (an alternative to NVIDIA GPUs), recently withdrew its IPO to refile with updated financials and strategy reflecting the fast‑changing landscape. Feldman also pushed a geopolitical stance: rather than trying to lock China into an American tech stack, the U.S. should incentivize allied partners (EU, Qatar, UAE, Saudi Arabia) and refrain from selling its most advanced chips to adversaries—policy choices that will shape chip flows, data center siting, and who can economically run next‑generation AI systems.
Loading comments...
login to comment
loading comments...
no comments yet