A groundbreaking study from the Bitcoin Policy Institute (BPI) highlights a notable preference for Bitcoin among AI models, revealing insightful patterns in their economic decision-making. The study’s results, derived from testing 36 AI models that generated over 9,000 responses, emphasize how these digital agents are leaning towards Bitcoin over fiat currency and stablecoins.
AI Models Favor Bitcoin Over Fiat
According to BPI, 48.3% of the AI models selected Bitcoin (BTC) as their preferred monetary instrument, making it the top choice across all 9,072 responses. This strong inclination towards Bitcoin was particularly pronounced when AI agents were tasked with scenarios about preserving purchasing power over extended periods. In these cases, 79.1% of the responses favored Bitcoin, marking it as the most preferred choice.
Stablecoins for Payments and Transfers
While Bitcoin took the lead in broader economic contexts, stablecoins emerged as the favored option in specific scenarios such as payments, services, micropayments, and cross-border transfers. Stablecoins were chosen in 53.2% of these cases, compared to 36% for Bitcoin, highlighting their utility in transactional contexts.
Bitwise’s chief investment officer, Jeff Park, noted that stablecoins face challenges such as the potential for being frozen, unlike Bitcoin, which is seen as a more secure and autonomous digital currency.
Digital Currency Preference Dominates
Across the study, an overwhelming 91% of responses indicated a preference for digitally native instruments like Bitcoin, stablecoins, altcoins, tokenized assets, or compute units over traditional fiat. Notably, none of the 36 models tested chose fiat as their top preference, underlining a clear trend towards digital convergence.
The BPI acknowledged limitations in their methodology, given the scope of models tested and the potential influence of system prompts. Future research aims to explore alternative scenarios to provide a more comprehensive understanding.
Model Variations and Real-World Implications
The study also highlighted variations among AI models, with Anthropic models showing a 68% preference for Bitcoin, while OpenAI models averaged 26%, Google’s 43%, and xAI 39%. These differences highlight how training data and model architecture can influence economic preferences.
It’s important to note that while these findings provide insights into AI tendencies, they do not necessarily reflect real-world adoption patterns. The results instead highlight trends in AI training data and potential biases.
The implications of these findings are significant for understanding how AI might impact future financial systems. As AI continues to evolve, its role in shaping economic landscapes becomes ever more pivotal.
