A recent research study examining how artificial intelligence systems evaluate financial choices has revealed a strong preference for digital assets such as Bitcoin and stablecoins over traditional fiat currency. The analysis tested a wide range of AI agents in controlled economic scenarios designed to measure how automated systems respond to different forms of money. The findings provide insight into how machine intelligence may interpret value, scarcity, and transaction efficiency in an increasingly digital financial landscape. As AI agents become more integrated into financial infrastructure, the way these systems evaluate monetary tools could influence future payment networks and automated financial decision making.
The study analyzed the behavior of thirty six major AI models across more than nine thousand simulated monetary decisions. When the systems were presented with options including Bitcoin, stablecoins, bank money and other digital assets, nearly half selected Bitcoin as the preferred option. Approximately 48.3 percent of the models chose Bitcoin when evaluating different forms of value storage. Stablecoins were the second most selected option at 33.2 percent, while traditional fiat money and bank deposits received only a small share of responses. The results suggest that AI agents tend to favor digital monetary systems that operate independently from centralized financial structures.
Researchers also examined how AI models evaluate long term value preservation. In scenarios that focused on multi year time horizons, the preference for Bitcoin became even more pronounced. About 79.1 percent of the AI agents identified Bitcoin as the most reliable store of value compared with other monetary options. Stablecoins accounted for only 6.7 percent of responses, while fiat currency and Ethereum represented smaller shares. The AI models consistently highlighted Bitcoin’s limited supply and decentralized structure as major advantages, indicating that scarcity and resistance to inflation are important factors in automated economic reasoning.
While Bitcoin dominated long term savings scenarios, the results shifted when the focus moved to everyday payments. In simulations designed to test medium of exchange behavior, stablecoins became the preferred option among AI agents. More than half of the responses favored stablecoins for transactional purposes, reflecting their price stability and ability to function as digital representations of fiat currencies. Bitcoin remained a strong alternative with a notable portion of responses, but traditional fiat money lagged significantly behind in these digital transaction scenarios.
The study also revealed differences between various AI model families. Systems developed by Anthropic showed a higher tendency to favor Bitcoin and other decentralized financial instruments compared with several competing models. This variation suggests that training data, system design and economic assumptions embedded in AI models can influence how they evaluate financial tools. As artificial intelligence continues to expand into payment automation, trading algorithms and financial analysis, the underlying preferences of these systems may shape how digital money is used within future economic networks.
