{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,28]],"date-time":"2026-10-28T10:42:57Z","timestamp":1793184177908,"version":"4.5.0"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":0,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>AI adoption can improve individual forecast precision while reducing the diversity of information across traders. We study this trade-off in a two-stage Kyle-type market where traders choose between idiosyncratic human signals and a more precise shared AI signal before trading. In both two-trader and continuum models, we identify conditions under which full AI adoption arises as the unique equilibrium yet yields lower combined expected profits for strategic traders compared to no AI adoption. This AI monoculture paradox arises because common forecast errors survive aggregation and synchronize order flow, allowing crowding to outweigh the gains from greater precision. In the continuum paradox regime, taxing AI adoption and returning the tax revenue through lump-sum rebates can align individual incentives with collective profit maximization, making the planner\u2019s optimal adoption rate the unique decentralized equilibrium. Using the theoretical model as a benchmark, we delegate information acquisition decisions to LLM agents in three sets of experiments examining best responses, beliefs about others, and communication. When other traders\u2019 adoption rate is given, agents can solve the individual optimization problem and choose the profit-maximizing signal. When forming their own beliefs about others\u2019 behavior, agents in partial-adoption settings often expect AI adoption to exceed the equilibrium rate. They then choose human signals to avoid anticipated crowding, driving actual adoption below equilibrium. With communication allowed, full adoption emerges even in a setting where theory predicts partial adoption; agents also attempt to mislead others for private gain.<\/jats:p>","DOI":"10.2139\/ssrn.7664018","type":"posted-content","created":{"date-parts":[[2026,10,28]],"date-time":"2026-10-28T10:02:47Z","timestamp":1793181767000},"source":"Crossref","is-referenced-by-count":0,"title":["The AI Monoculture Paradox in Financial Markets"],"prefix":"10.2139","author":[{"given":"Jose","family":"Blanchet","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chenghan","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renyuan","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,10,28]],"date-time":"2026-10-28T10:02:47Z","timestamp":1793181767000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=7664018"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":0,"URL":"https:\/\/doi.org\/10.2139\/ssrn.7664018","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}