Three years ago, Sequoia partner David Cahn estimated that the AI industry needed to earn $3 trillion by 2026 to justify the massive spending on infrastructure. Fast-forward and the number has jumped to $1.5 trillion—a hefty sum indeed.
In this financial landscape, companies like Anthropic and OpenAI are racking up impressive revenues but still fall short of Cahn’s ambitious targets. Meanwhile, Torsten Slok at Apollo warns that if tech giants don’t meet their projected free-cash flow in 2028, it could send shockwaves through the economy.
The challenge? Cheaper open-source AI models are becoming more popular, potentially reducing token usage and revenue streams for established players. This shift could mean cheaper access to AI but also a crunch for companies banking on higher token efficiency.
As we herd our AI agents toward cheaper tokens, it’s clear that the future of this technology hinges not just on innovation but on financial acumen. The risk is real: miss the mark and you might see the S&P 500 face its own correction.
