On Oct. 31, the Financial Stability Board, the official body that coordinates financial regulation among the G-20 nations, published a report on the growing use of artificial intelligence and machine learning in financial services. The FSB report said the use of these technologies can provide benefits through the more efficient processing of information, but also identified several risks to financial stability. These include new and unexpected forms of interconnectedness, third-party dependencies, and the difficulties in interpreting and auditing AI and ML methods and models.
For example, many models are being "trained" in a period of low volatility, the report noted. "As such, the models may not suggest optimal actions in a significant economic downturn or in a financial crisis, or the models may not suggest appropriate management of long-term risks."
Another potential risk is dependence on a handful of technology firms that dominate the provision of these services to the financial sector. "These third-party dependencies and interconnections could have systemic effects if such a large firm were to face a major disruption or insolvency," the report warned.