Jensen Huang resigns as Nvidia CEO; CFO Colette Kress named interim chief

Jensen Huang's exit fractures the "personalized moat" that underpinned CUDA's ecosystem dominance. For 33 years, developer trust, TSMC advanced-node scheduling, and hyperscaler custom-silicon negotiations were anchored to his individual technical judgment and relationship capital. The board's selection of a CFO over a CTO signals a decisive governance pivot: from technology-narrative-driven expansion to capital-discipline-first stewardship. Technically, CUDA's stickiness was never purely a software barrier—it was built on Huang's direct, sustained engagement with top-tier researchers and architects. The transition window hands AMD's ROCm, Google's TPU, and AWS Trainium a rare strategic opening. Hyperscalers will accelerate multi-vendor diversification, compressing single-source exposure to historic lows within two quarters. On compliance, an interim CFO structure inevitably slows sovereign AI infrastructure deals, which demand CEO-level speed, cross-timezone negotiation, and geopolitical acuity. A finance-background leader carries an inherent response latency in these high-stakes engagements. The 12-to-24-month long-tail: expect a structural "execution discount" on Nvidia's valuation, shifting the pricing logic from growth premium to earnings certainty—echoing the post-Jobs Apple compression of 2011. Yet CUDA's developer inertia and migration costs will sustain the core business for roughly 18 months. The true inflection point arrives with the 2026 Rubin architecture. Whether it ships on schedule will determine whether this transition becomes a managed generational handoff or a fundamental strategic reset.