Not a professor, but my under-credentialed hot take:
- The "decline" narrative is a feature of every technology S-curve. We're not at the top. We're still near the bottom.
- OpenAI and Anthropic and others will keep competing, and that competition will keep them innovating (declaring the race over is the oldest mistake in tech).
- Innovations won't just come from model performance. The real differentiator is the architecture layer that turns model power into things that actually get done. Everyone will figure this out, and "harness" will become the word of the decade.
- Google is better positioned than any other competitor to integrate AI into real enterprise workflows.
- The companies that built hard infrastructure (data centers, chips, power) aren't going anywhere. Nvidia's customers are building things that will take years to complete. Demand doesn't evaporate because a pundit says margins are thinning.
- Chinese AI companies will compete. They'll also hit their own walls. Assuming they win by default is as naive as assuming they lose by default.
- LLMs becoming commodities is not the death of AI value. Think of cloud compute. Amazon, Microsoft, and Google didn't collapse when servers got cheap, they got richer because cheap infrastructure unlocked entirely new categories of applications.
- The agentic layer is where the next fortunes get made. Not in the model itself, but in what the model can orchestrate, automate, and extend. We're barely in the first inning.
- SpaceX's AI efforts will be fine. Dismissing a company with the best vertical integration in aerospace and a workforce that ships hardware on compressed timelines seems premature.
- Human-AI teaming will define economic and national security outcomes over the next decade in ways that dwarf any previous technological shift. The organizations that figure this out first won't be the loudest ones on X.
It is so good to be alive in the age of early stage capitalism.
Hot take on what comes next, after the sudden decline of tokenmaxxing:
- OpenAI will struggle
- with the decline of tokenmaxxing Anthropic will struggle (aside from this quarter) to make a profit
- Google will catch up to Anthropic
- some Chinese companies might, too
- LLMs will become commodities; margins will be very very thin
- Most of the companies that invested massively in them will struggle to make back their investments
- SpaceX’s AI efforts will flail
- Nvidia will eventually decline, once all of the above becomes widely recognized.