While the West obsesses over algorithmic superiority, the true AI battlefield is physical infrastructure. China's dominance in manufacturing data center components and its potential to compromise the power grid represent a more fundamental strategic threat than model capabilities.
The discourse often presents a binary: AI plateaus below human level or undergoes a runaway singularity. A plausible but overlooked alternative is a "superhuman plateau," where AI is vastly superior to humans but still constrained by physical limits, transforming society without becoming omnipotent.
Unlike math or code with cheap, fast rewards, clinically valuable biology problems lack easily verifiable ground truths. This makes it difficult to create the rapid reinforcement learning loops that drive explosive AI progress in other fields.
The FDA approved Artera AI’s prostate cancer diagnostic without understanding *why* it works. This precedent suggests that massive retrospective validation on patient data can substitute for model interpretability, changing the strategic focus for medical AI companies.
Many leaders at frontier AI labs perceive rapid AI progress as an inevitable technological force. This mindset shifts their focus from "if" or "should we" to "how do we participate," driving competitive dynamics and making strategic pauses difficult to implement.
Key decisions during data center construction, like granting personnel access to site plans, are "one-way doors." Once a potential adversary has this information, the compromise is baked in, and the facility's security cannot be fully restored later.
International AI treaties, particularly with nations like China, are unlikely to hold based on trust alone. A stable agreement requires a mutually-assured-destruction-style dynamic, meaning the U.S. must develop and signal credible offensive capabilities to deter cheating.
A critical, under-discussed constraint on Chinese AI progress is the compute bottleneck caused by inference. Their massive user base consumes available GPU capacity serving requests, leaving little compute for the R&D and training needed to innovate and improve their models.
China's rise in biotech isn't just about cost. It's driven by a tightly integrated ecosystem where drug designers and wet lab technicians work closely, creating a much faster feedback loop than the siloed, outsourced model common in the US.
Even if AI accelerates parts of a workflow like coding, overall progress might stall due to Amdahl's Law. The system's speed is limited by its slowest component, meaning human-dependent tasks like strategic thinking could become the new rate-limiting step.
AI adoption isn't linear. A small, 1% improvement in model capability can be the critical step that clears a usability hurdle, transforming a "toy" into a production-ready tool. This creates sudden, discontinuous leaps in market adoption that are hard to predict from capability trend lines alone.
Power users are building personal AI assistants not just by feeding data, but by creating curated context layers. This involves exporting all digital communications (email, Slack), then using LLMs to create tiered summaries (e.g., monthly chief-of-staff briefs) to give agents deep, usable context.