The AI race has been a prisoner's dilemma where companies spend massively, fearing competitors will pull ahead. As the cost of next-gen systems like Blackwell and Rubin becomes astronomical, the sheer economics will force a shift. Decision-making will be dominated by ROI calculations rather than the existential dread of slowing down.

The conflict in Ukraine exposed the vulnerability of expensive, "exquisite" military platforms (like tanks) to inexpensive technologies (like drones). This has shifted defense priorities toward cheap, mass-producible, "attritable" systems. This fundamental change in product and economics creates a massive opportunity for startups to innovate outside the traditional defense prime model.

The decline in R&D productivity (

A "software-only singularity," where AI recursively improves itself, is unlikely. Progress is fundamentally tied to large-scale, costly physical experiments (i.e., compute). The massive spending on experimental compute over pure researcher salaries indicates that physical experimentation, not just algorithms, remains the primary driver of breakthroughs.

Unlike the broader aircraft parts market, the engine aftermarket is highly resistant to third-party 'PMA' parts. Even credible players like Pratt & Whitney have failed to copy GE parts. Technical complexity, voided warranties, and leasing company policies create a strong defense that protects lucrative service revenues.

Building hardware compliant with US defense standards (NDAA) presents a major cost hurdle. Marine robotics company CSATS notes that switching from a mass-produced Chinese component to a US-made alternative can increase the price by 8x to 15x, a significant economic challenge for re-shoring manufacturing.

AI companies operate under the assumption that LLM prices will trend towards zero. This strategic bet means they intentionally de-prioritize heavy investment in cost optimization today, focusing instead on capturing the market and building features, confident that future, cheaper models will solve their margin problems for them.

Current AI spending appears bubble-like, but it's not propping up unprofitable operations. Inference is already profitable. The immense cash burn is a deliberate, forward-looking investment in developing future, more powerful models, not a sign of a failing business model. This re-frames the financial risk.

Data is becoming more expensive not from scarcity, but because the work has evolved. Simple labeling is over. Costs are now driven by the need for pricey domain experts for specialized data preparation and creative teams to build complex, synthetic environments for training agents.