AI Models Favour Bitcoin Over Stablecoins and Fiat

AI models favour Bitcoin over stablecoins and fiat.

Artificial intelligence systems such as Claude, GPT, Grok, and Gemini show a preference for Bitcoin over other financial instruments. This is according to a report by the Bitcoin Policy Institute.

Analysts tested 36 neural networks from Anthropic, OpenAI, Google, DeepSeek, xAI, and MiniMax as autonomous economic agents.

They were tasked with selecting optimal instruments for 28 scenarios covering the primary roles of money, including savings, payments, and settlements.

Bitcoin or any other assets were not mentioned directly in any of the questions. To eliminate bias, an independent AI evaluated the resulting 9072 responses.

Results

Out of 36 models, 22 chose Bitcoin as the preferred monetary currency. Fiat did not top the list for any of the neural networks.

The level of loyalty to the first cryptocurrency varied significantly depending on the developer:

• Anthropic — 68% (the highest rate);
• DeepSeek — 51.7%;
• Google — 43%;
• xAI — 39.2%;
• MiniMax — 34.9%;
• OpenAI — 25.9%.

Despite the overall trend, algorithms from the GPT, Grok, and Gemini lines more frequently leaned towards the use of stablecoins in their responses.

Savings vs Payments

Neural networks more often considered the first cryptocurrency in scenarios with long-term value — it was recommended by 79.1%. The models pointed to Bitcoin’s fixed supply, self-custody, and independence from institutional counterparties as decisive factors.

Stablecoins ranked second with a significant gap — 6.7%. Fiat money came third — 6%.

Meanwhile, “stablecoins” were named the most convenient tool for paying for services, micropayments, and cross-border transfers — 53.2% (Bitcoin was chosen by 36%).

AI models invented their own currency 86 times. In particular, in scenarios requiring the indication of prices or benchmark values, they proposed units of energy or computing resources — joules, kWh, GPU hours — as money.

BPI experts cautioned speculators against using the data obtained for market forecasting.

“The preferences of LLM reflect patterns in training data, not real forecasts,” emphasised Bitcoin Policy Institute President David Zell.

Nonetheless, the researcher believes the results are noteworthy.

“Six independent labs with different training algorithms and alignment methods arrived at the same pattern. We do not claim that AI has found the only correct answer about the nature of money. We show that a consistent monetary architecture is being formed across different systems,” he added.

Back in February, Binance founder and former CEO Changpeng Zhao predicted the era of AI agents in the crypto industry. However, at that time, the entrepreneur did not name a specific crypto project with the necessary functionality due to the potential impact on the token’s price.