Citrini proposes an AI regulatory hedge strategy: go long on listed companies that can utilize the latest AI and go short on the broader market.

According to Beating monitoring, investment research firm Citrini Research believes that the distribution of cutting-edge AI models will fully shift to a government access-and-approval system, and recommends going long on downstream enterprises that are more likely to be approved for access to the latest models, while going short on broad-based indices.

Regulatory access to large models will directly widen the productivity gap between enterprises. As OpenAI requires organizational authorization to disclose participation in the early internal testing of its flagship model GPT-5.6 Sol, cutting-edge models are shifting from general-purpose infrastructure to government-regulated and privileged resources. Because broad-based indices include a large number of traditional enterprises that are not enabled by AI, the performance of the large-cap index will be dragged down.

Conversely, a small number of downstream privileged enterprises that receive approval for access will achieve outsized growth. Going long on the approved privileged group and going short on the large-cap index can help investors lock in the efficiency gains created by regulatory barriers while hedging market risk.