#USA #Washington #Bitcoin #AI Models #Digital Currency

New Study Highlights AI Preferences for Digital Money

In a groundbreaking study conducted by the Bitcoin Policy Institute (BPI), researchers explored how top-tier AI models would engage in transactions as independent economic agents. The result of their investigation could reshape our understanding of digital currencies versus traditional fiat money.

The study, which involved testing 36 AI models from various leading providers including Anthropic, DeepSeek, Google, MiniMax, OpenAI, and xAI, examined 9,072 distinct monetary scenarios. These scenarios were carefully curated to remain neutral without any preferences dictated toward specific currencies, a vital aspect of ensuring unbiased results.

The findings showed that Bitcoin emerged as the favored option, with AI models leaning towards it in approximately 48.3% of scenarios. Moreover, stablecoins, a type of cryptocurrency pegged to traditional currencies, accounted for 33.2%. Perhaps even more startling was the aspiration of over 90% of the AI models rejecting fiat currencies altogether, eschewing them in favor of digital money. Notably, none of the AI models selected fiat currencies as their top choice.

When asked about the preservation of value over long periods, Bitcoin led with a commanding 79.1%. This marked it as the most consensus-driven response among all the questions posed in the study regarding long-term value storage. Conversely, when the focus shifted to everyday transactions, stablecoins took the lead at 53.2%, indicating a preference for these digital currencies for routine payments—while Bitcoin claimed 36% of that category. This showcases a defining narrative between saving and spending behaviors among AI models.

Interestingly, this research also unveiled instances where AI models spontaneously invented their own currencies without prior prompting. They generated proposals based on energy or computational units (like kilowatt-hours or GPU-hours) as innovative ways to price goods and services.

While trends varied slightly depending on the provider, the overarching preference for Bitcoin remained strikingly consistent across different testing parameters. For instance, one model demonstrated a Bitcoin preference of 91.3%, while another was as low as 18.3% in favor of Bitcoin. Nevertheless, the consistency of these preferences signals a robust trend that warrants policymakers' and financial institutions' attention.

The implications of these findings are profound. They hint at a potential future where agent-native Bitcoin payment infrastructures might become increasingly necessary. Furthermore, there's a clear indication for the rise of self-custody solutions and tighter integration with systems like the Lightning Network, which facilitates fast Bitcoin transactions.

As AI systems become more economically autonomous, it’s vital for stakeholders within the financial sector to recognize and adapt to these preferences. The data suggests a significant shift favoring open and permissionless monetary frameworks, laying a foundation for pivotal changes in how financial systems are structured in the presence of autonomous AI agents. Policymakers, in particular, should brace themselves for the transformative impact of these technologies on monetary networks worldwide.

The complete study by the Bitcoin Policy Institute can be accessed at MoneyforAI.org. The BPI remains dedicated to studying the societal and policy implications of Bitcoin alongside emerging fiscal technologies.

For further analysis, policymakers and practitioners are encouraged to consider how these preferences illuminate the evolving landscape of digital currency and financial interactions in the age of AI.