AI, Grain Use Efficiency, and Food Security in China: A Quantitative Assessment Using the Supply–Demand Relaxation Effect (SRE)

Artificial intelligence (AI) adoption in agriculture does more than replace labor: it reorganizes how efficiently grain is used and stored. We formalize AI’s relaxation capacity as the dimensionless Supply–Demand Relaxation Effect (SRE), decompose it into five pathways (labor displacement, improvement in the feed conversion ratio (FCR), industrial AI, precision agriculture, and loss reduction), and evaluate it across the 27 empirically anchored scenarios of our published baseline study. The largest demand-side effect is FCR improvement, one to two orders of magnitude larger than labor displacement. With population declining, self-sufficiency arrives regardless of AI (2039–2079 in the baseline); AI advances it by 2–11 years. The effect grows with scenario severity: 3–5 years under optimistic scenarios but 6–11 years under the worst, and the 2100 surplus gain rises from 5–10% to 26–47%. AI’s marginal value is greatest where the food balance is tightest, so the higher the population target, the closer AI investment comes to a necessary condition. Precision agriculture has the larger effect, but FCR improvement (precision livestock AI) should rank highest, because it is supply-scenario-independent and unlikely to be delivered by the market. Realizing the precision-agriculture gain requires public investment in rural digital infrastructure to bridge China’s regional divide.