AI-Driven Trading: Reducing Latency and Human Bias in Financial Markets
摘要
Following the advent of artificial intelligence, the financial market operates at a technological speed and is subject to human fallibility. This paper addresses latency, defined as the time between when an order is initiated and executed, and its impact on trading outcomes. Technological speed improves efficiency but introduces systemic vulnerabilities when paired with human biases, such as overconfidence, herding, and loss aversion. Using three case studies, a new discernment framework tool is proposed as an innovative governance process to complement current quantitative methods and reduce the risks associated with AI-driven latency. The framework combines scenario analysis, a literature review, and a comparative evaluation of latency strategies, highlighting the challenges and opportunities associated with increased trading speed. This conceptual paper reminds us of the importance of maintaining an ethical leadership and governance framework in the financial market, and how discernment can help minimize issues.