The abrupt departure of Fermi Inc. co-founder and CEO Toby Neugebauer on April 17, 2026, along with related changes involving CFO Miles Everson, sent shares of the AI-focused data center developer plunging as much as 30%, underscoring mounting operational and financial pressures at one of the sector's most ambitious projects. Fermi, which went public in 2025 with plans for Project Matador—an 11 GW+ energy and data center campus in the Texas Panhandle near Amarillo powered initially by natural gas and eventually nuclear reactors—has struggled to secure anchor tenants despite the purported scarcity of AI-ready power. A major $150 million tenant deal collapsed in December 2025, contributing to a $486 million net loss for the year and leaving the company without committed revenue streams for its REIT structure. This is not an isolated setback: the company has faced a class-action lawsuit alleging overstated tenant demand, while insiders have been selling millions in shares as the stock has cratered over 80% from its post-IPO highs. Analysts noted the management overhaul, which installed an interim Office of the CEO under COO Jacobo Ortiz and Anna Bofa, may ease customer negotiations, but it highlights friction at the highest levels after months of board deliberations. Mainstream coverage often frames these events as company-specific execution issues, yet they align with broader warnings of energy bottlenecks threatening the hyperscale AI narrative. Nearly half of planned U.S. data center builds in 2026 have been delayed or canceled due to shortages of power infrastructure, transformers, and grid connections—exacerbated by reliance on imports and competing demands from EVs and electrification. Forecasts indicate U.S. data center power demand could hit 74 GW by 2028 against a projected 49 GW shortfall in available supply in the near term, with short-term risks of overbuild in commodity capacity even as AI-optimized sites remain constrained. Fermi's pivot toward nuclear, while visionary, faces the same long-lead timelines and regulatory hurdles dogging the industry. These dynamics suggest the 'AI power scarcity' story—promoted heavily by hyperscalers—may mask a bubble where capital expenditure projections exceeding $3 trillion through 2030 outrun realistic infrastructure delivery. Downplayed in much coverage are the toxic feedback loops: without signed tenants, financing stalls; without proven power delivery, tenants hesitate. Fermi's troubles, including its shift to a new Dallas headquarters, serve as a canary for sector-wide strains where euphoria around exponential AI growth collides with physical limits on energy and equipment. While some analysts like those at Morgan Stanley see long-term demand justifying investments, the near-term pain at pioneers like Fermi points to consolidation, delayed projects, and potential write-downs that could reshape the AI infrastructure landscape by 2027.