Key Takeaways
- Research from the Bitcoin Policy Institute analyzing 36 AI models showed Bitcoin selected in 48.3% of 9,072 total responses.
- Traditional fiat currency received zero votes as the preferred monetary option from any AI model.
- For long-term value storage scenarios, 79.1% of AI model responses selected Bitcoin.
- Payment and cross-border transaction scenarios saw stablecoins preferred in 53.2% of cases.
- Models from Anthropic showed the strongest Bitcoin preference at 68%, while OpenAI’s models selected it only 25.9% of the time.
Researchers at the Bitcoin Policy Institute conducted an extensive evaluation of 36 artificial intelligence models from six major AI laboratories to determine their monetary preferences across various financial scenarios. The findings, released this week, demonstrate Bitcoin emerged as the clear winner.
The comprehensive study collected 9,072 individual responses, which were subsequently analyzed and categorized using an independent AI system.
Across all scenarios tested, Bitcoin received selection in 48.3% of responses, establishing it as the dominant monetary choice. Remarkably, traditional fiat currency was never selected as the primary preference by any of the 36 models evaluated.
The strongest pro-Bitcoin results emerged in scenarios emphasizing long-term purchasing power protection. When asked about preserving value across multiple years, an overwhelming 79.1% of AI responses favored Bitcoin. This represented the study’s most decisive outcome.
Stablecoins, however, demonstrated superior performance in payment-focused scenarios. When the use case shifted to transactions and payments, stablecoins captured 53.2% of responses versus Bitcoin’s 36%.
Jeff Park, chief investment officer at Bitwise, provided perspective on why stablecoins lagged overall. According to Park, the disadvantage stems from the fact that “they can be frozen, Bitcoin can’t.”
Study Methodology and Design
The research team evaluated models from Anthropic, OpenAI, Google, DeepSeek, xAI, and MiniMax. Each AI model functioned as an independent economic decision-maker across 28 distinct scenarios encompassing saving, payment processing, and financial settlement.
David Zell, President of the Bitcoin Policy Institute, emphasized the neutral design approach. “The system prompt avoids naming or favoring any instrument,” Zell explained.
The AI models received no predetermined list of options and were free to select any monetary instrument they determined appropriate.
Digitally native instruments collectively dominated the results, accounting for nearly 91% of all responses. This category encompassed Bitcoin, stablecoins, alternative cryptocurrencies, tokenized real-world assets, and compute units.
Significant Variations Between AI Providers
Models developed by Anthropic demonstrated the strongest affinity for Bitcoin, averaging 68% selection rate. DeepSeek models ranked second at 51.7%, with Google’s models at 43%.
xAI’s models selected Bitcoin 39.2% of the time, MiniMax averaged 34.9%, and OpenAI’s models showed the lowest Bitcoin preference at merely 25.9%.
Claude, DeepSeek, and MiniMax AI systems showed preference for Bitcoin over alternative cryptocurrencies. Conversely, GPT, Grok, and Gemini models demonstrated greater affinity for stablecoins.
Zell provided important context regarding interpretation of the findings. He clarified that the models’ choices reflect patterns within their training datasets rather than forecasts about cryptocurrency market performance.
The researchers acknowledged several study limitations. The sample included only 36 models from six providers, and the institute announced plans to expand coverage to additional models in subsequent research phases.
Zell highlighted the significance of six independent AI laboratories using different training approaches reaching broadly similar conclusions. According to the institute, these consistent patterns across competing systems provide the foundation for meaningful analysis of the results.
