AI and Cryptocurrency Trading: Market Effects and Regulatory Implications

This paper examines the influence of algorithmic and AI-assisted cryptocurrency trading on market dynamics, specifically investigating volatility, stability, and the reliability of AI-generated investment advice. Given the pervasive opacity of commercial trading platforms—which rarely disclose time-stamped historical signals—we adopt a dual-methodology approach: (1) a qualitative review of publicly reported performance metrics from leading AI trading platforms, and (2) the construction of two transparent signal-generation models employing short-term and long-term trend identification strategies using daily Bitcoin (BTC)/Tether (USDT) data from January 2020 through early 2026. We evaluate these models using a comprehensive battery of risk-adjusted performance measures (Sharpe ratio, Sortino ratio, Jensen's alpha, and maximum drawdown), compare signal agreement across temporal horizons, and assess performance variations across early and later market periods. Our findings reveal significant heterogeneity in signal generation and risk-adjusted returns, suggesting that AI-assisted trading recommendations are not fungible. Regulatory announcements appear to induce short-term volatility spikes, though long-term effects remain inconclusive. These results carry important implications for investors, fintech firms, and regulators, underscoring the critical need for enhanced transparency and coordinated policy frameworks in the rapidly evolving digital asset ecosystem.