AI Data Center Power Bottleneck Makes Bloom Energy This Analyst's Top 2026 Stock Pick
Beth_Kindig · x · 2026-09-19
Beth Kindig argues the real risk to the AI economy isn't compute or software but power availability — specifically time-to-power. Key data: Big Tech is spending $650B+ annually on AI data centers; NVIDIA's GB200/GB300 racks draw 120-140kW (2x H200's 70kW); rack-scale systems will need 300-600kW within 1-2 years, a 5x jump from H1 2025. Grid and nuclear expansion are too slow, so her top 2026 pick is Bloom Energy, whose solid oxide fuel cells provide behind-the-meter on-site generation to bring new data centers online faster ($BE, $NVDA).
More from Venture
- Gary Marcus: AI data center demand has already passed its peak as projects get killed — GaryMarcus · 2026-11-04
- Gary Marcus: 'fleecing the public with AI fantasies just got harder' as AI stocks drop — GaryMarcus · 2026-11-04
- Dev's AI agents auto-submit his product to directories, 4 confirmed in 16hrs — pramodk73 · 2026-11-04
- Reddit case study: agent discovery optimization brought $14k MRR in a month — tomarv99 · 2026-11-04
- Indie dev hits $25k MRR in one year after 7 years of failures — rudrank · 2026-11-04
- Chinese AI stocks MiniMax and Zhipu pump 15% in an hour — menhguin · 2026-11-04