With only six months of runway, NVIDIA couldn't afford a typical hardware development cycle. CEO Jensen Huang spent half their remaining capital on an unproven emulator from a failed company. This high-risk bet on process innovation was the only way to test their comeback chip before manufacturing.

Despite investing $80B, Meta is shuttering its metaverse project. This avoids the sunk cost fallacy—the irrational commitment to a failing venture based on past investment. The smart move is to cut losses and reallocate future resources to more promising areas like AI.

Nvidia invests broadly in AI startups because of its own origin story. Surviving as one of 63 graphics companies despite having a "precisely wrong" architecture taught CEO Jensen Huang the folly of trying to pick winners in a nascent market.

Facing bankruptcy after canceling a pivotal contract with Sega, CEO Jensen Huang persuaded Sega's CEO to convert the remaining $5M payment into an investment. This act of radical candor and persuasion, done without a clear business plan, provided the capital necessary for survival.

Facing bankruptcy in the 90s, NVIDIA couldn't afford to build a physical prototype for its make-or-break NV3 chip. The team relied entirely on simulation, a high-risk strategy that paid off and saved the company, ironically foreshadowing its future dominance in creating hardware for complex simulations.

During a routine roadmap review, Nvidia's CEO unexpectedly abolished a major product line and reassigned a third of the company's engineers. This exemplifies the fearless, rapid, and decisive leadership required to navigate fast-moving tech markets.

To solve the chicken-and-egg problem for its CUDA platform, NVIDIA included the costly technology in every gaming GPU sold. This knowingly depressed margins for over a decade but created a massive installed base that eventually attracted the researchers who kickstarted the AI revolution.

Despite just launching its first-generation autonomy system, Rivian completely reset it, throwing away all the code and hardware. CEO RJ Scaringe said the decision was easy because it was obvious that the old rules-based architecture had a 0% chance of being competitive against modern neural net-based approaches.

Contrary to chatter that suggests OpenAI is "flailing" by killing multiple high-profile products, this is a sign of strong business discipline. Aggressively avoiding the sunk cost fallacy allows the company to pivot resources to core priorities like enterprise sales, which is a long-term strategic strength.