Global financial regulators are urging for stricter controls on the use of increasingly autonomous artificial intelligence systems within the financial sector, as adoption accelerates. The Financial Stability Board (FSB), a key international standard-setter, released a report on Wednesday emphasizing the potential for these AI systems, particularly 'agentic' AI capable of planning and executing tasks with minimal human intervention, to amplify risks.
Agentic AI is already being deployed by financial firms for tasks such as fraud detection, customer service, and back-office operations. A survey by the Cambridge Centre for Alternative Finance found that 52% of financial sector respondents reported active adoption of agentic AI, with 23% scaling or transforming their use and 29% piloting these functions. Regulators and global bodies have intensified warnings about AI's risks since Anthropic released Mythos, which experts view as presenting significant cybersecurity challenges to the banking industry.
The FSB report highlights that autonomous AI introduces risks that can materialize rapidly, including the potential for unauthorized or illegal actions, data breaches, and disruption to interconnected systems. The report specifically notes that AI agents pose a distinct challenge to human oversight, as they might pursue actions deviating from firm intentions without staff awareness or the ability to intervene promptly.
To mitigate these risks, the FSB has outlined proposed "sound practices." These non-binding guidelines, open for feedback until July 22, urge financial firms to establish clear boundaries for AI use and embed safeguards. Recommendations include defining the scope of AI agent actions and mandating human approval for high-risk activities, such as financial transactions exceeding certain thresholds. The FSB also suggested adapting HR controls to treat AI agents as "synthetic employees."