Singapore's government has built a dating app called FirstDate that runs on the Gale-Shapley deferred acceptance algorithm, the stable matching mechanism developed in 1962 that won Lloyd Shapley and Alvin Roth the 2012 Nobel Memorial Prize in Economic Sciences. The same algorithm currently underpins the US National Resident Matching Program for hospital placements and kidney exchange networks. The app is currently restricted to public servants aged 21-35. The mechanism works in rounds. Users' preferences and dealbreakers generate a ranked priority list. Proposers offer to their top-ranked choice; receivers hold their best offer and reject the rest. Rejected proposers move down their list. The cycle repeats until every participant is matched. The mathematical guarantee: the resulting matching is stable, meaning no two unmatched people exist who would both prefer each other over their assigned partners. The UX design enforces scarcity and commitment at every layer. Users receive one match per cycle with zero infinite scroll. A 72-hour decision window forces action. Contact information is revealed only on mutual acceptance. Identity is verified through Singpass, Singapore's national digital identity system, eliminating catfishing by design. The structural inversion from commercial dating apps is the core story. Tinder, Hinge, and Bumble are advertising businesses whose revenue depends on continued engagement — they need you swiping. FirstDate's success metric is the opposite: users deleting the app because they found a partner. Singapore has built what may be the first dating platform whose institutional incentive is aligned with the user's actual goal. There is a well-known asymmetry in Gale-Shapley worth noting. The algorithm yields proposer-optimal, receiver-pessimal outcomes — proposers get the best stable match available to them, while receivers get their worst stable partner among all possible stable matchings. Which side proposes matters enormously, and how Singapore assigns proposer vs. receiver roles will shape outcomes in ways most users will never see. The restriction to public servants aged 21-35 makes this a controlled pilot, not a population-scale intervention. Singapore's fertility rate hit 0.97 in 2023, the lowest recorded, and pro-natalist policy is an explicit government priority. Whether game-theoretic matching can move demographic needles that cash incentives have not is an open question. The app is a policy experiment dressed as a product. The deeper question is whether state-run matching can outperform market-driven engagement optimization when the state's objective function — stable pairings — is fundamentally different from the market's — maximized screen time. If it works, expect copycats. If it doesn't, the failure mode will be instructive: either the algorithm isn't the bottleneck in human pairing, or the restricted user pool is too small for Gale-Shapley to find good matches.
