The discourse on AI's cultural missteps is a distraction. The correct strategic framework is viewing it as an arms race for intelligence between the US and China, which recontextualizes the massive spending and urgency as matters of national security and dominance.
While AI companies are taking on significant debt, it represents only about 4% of their total value. This is a fraction of the 30% debt-to-value ratio seen at the height of the dot-com mania, suggesting the current financial situation is far more stable and less leveraged than that historical bubble.
Challenging standard accounting, Nvidia's CEO argues that AI data centers are not rapidly depreciating assets. He points to six-year-old H100 chips still in use and notes that the cost of compute per hour has actually risen. He is even offering to financially backstop the value of his chips against decline.
China's AI strategy is not to beat the US on building the most advanced "frontier" models, but to create "good enough" open-source alternatives that are significantly cheaper. This price war threatens to hollow out the revenue of US AI leaders, even if US technology remains superior.
Hyperscalers like Google are spending from massive profits, not because their core business is failing. This AI expenditure is an "optional" bet, similar to Meta's Metaverse pivot. They can stop at any time and revert to being cash cows, a key distinction from fundamentally unprofitable ventures.
The narrative that companies regret AI-driven layoffs is misleading. They are replacing employees resistant to AI with new talent possessing the same job title but a different, AI-native skill set. It's a workforce transformation disguised as rehiring for roles that now require AI proficiency.
Because AI is framed as a systemic national security issue—an arms race with China—the US government is unlikely to let the industry fail. Individual companies might stumble, but the government will likely step in to prop them up or facilitate asset sales, creating an implicit taxpayer-funded backstop.
A key paradox exists in the AI market. While corporate debt is manageable (4% of value), the stock market's CAPE ratio is at 40x, a level last seen right before the 2000 dot-com crash. This suggests that while companies are financially sound, investor valuations may be dangerously over-extended.
A contrarian view on "destructive scanning" of old books for AI training. Instead of a loss, it's a form of preservation. This process transfers knowledge from a format no longer widely consumed (physical books) into a dynamic, interactive system (AI), making the information more accessible to the world.