Скачать или смотреть The $1.7 Trillion Energy Lie Behind Every AI Data Center

The AI bubble has a physics problem that no amount of venture capital can solve. Everyone's focused on NVIDIA GPU shortages, but GPU chips are a data center procurement problem. You write a check, you wait, they show up. Energy infrastructure is a physics problem, and it doesn't scale on Silicon Valley timelines.

Get 20% off of Magic Mind: https://www.magicmind.com/DRJOSH20

In this video I break down:
→ Why electricity is the true bottleneck in the AI stack (not GPUs, not software)
→ The actual numbers: 200GW of new capacity needed by 2030
→ How a single Meta data center draws more power than I use in my entire metro area
→ Why nuclear can't save us (despite what Sam Altman and Bill Gates want you to believe)
→ The company branding itself like a tech startup that should make you nervous
→ What Cornelius Vanderbilt can teach you about who actually wins in a tech boom

Sources: IEA Global Energy Outlook, JLL 2026 Global Data Center Outlook, U.S. Department of Energy estimates, Nuclear Regulatory Commission filings.