In AI for science, the true competitive advantage lies in generating unique, high-quality experimental data from self-driving labs. The AI models themselves are becoming commoditized, while the physical data remains the defensible asset.

Unlike digital AI trained on public internet data, physical AI models require vast, private datasets collected from real-world operations like mines. The ability to collect this proprietary data, often in restricted locations with government approval, creates a powerful and defensible competitive advantage.

Unlike consumer AI trained on public internet data, industrial AI requires vast, proprietary datasets from the physical world (e.g., sensor readings from a submarine hull). Gecko Robotics is building this data corpus via its robots, creating an advantage that's difficult to replicate.

China's scale in physical hardware like drones and autonomous vehicles generates a vast dataset of multimodal data (sound, vision, LiDAR). This real-world data, underappreciated in the text-focused West, gives Chinese companies a significant advantage in training intelligent physical AI models.

As AI and better tools commoditize software creation, traditional technology moats are shrinking. The new defensible advantages are forms of liquidity: aggregated data, marketplace activity, or social interactions. These network effects are harder for competitors to replicate than code or features.

In the AI era, models and compute infrastructure have near-zero switching costs and are becoming commoditized. A company's unique, historical data is emerging as its most valuable and defensible asset. This proprietary data, once archived and ignored, is now the key differentiator and competitive moat.

As AI commoditizes software, hardware is re-emerging as a key defensibility layer for startups. A decade ago, VCs avoided hardware, but now a physical device tied to a software subscription creates powerful stickiness and justifies high valuations, representing a major shift in investment strategy.

AI makes software incredibly easy to build and replicate, eroding traditional business moats. Chip Huyen argues the next frontier for durable value is in physical AI and robotics, where hardware development cycles and real-world complexities prevent instant copying.

As AI application layers become easier to clone, the sustainable competitive advantage is moving down the tech stack. Companies with unique, last-mile user interaction data can build proprietary models that are cheaper and better, creating a data flywheel and a moat that is difficult for competitors to replicate.