FSB strengthens recommendations on oversight of agentic AI — Channel NewsAsia

In London, the Financial Stability Board (FSB) released final recommendations for banks and other financial institutions on managing the risks of agentic artificial intelligence systems. The organization called for stronger human oversight of such systems, Channel NewsAsia reports.

Oversight for high-risk operations

The FSB said that more intensive human control is critically important in high-risk applications. Companies were advised to establish clear risk-based triggers under which an AI agent must seek human approval.

Agentic AI systems can plan and carry out tasks with limited human oversight. In its June draft recommendations, the FSB warned that this creates distinct challenges for human control.

More current news is available on the UA.News Telegram channel Telegram.

Changes in the final recommendations

In the final version of the document, the FSB strengthened its recommendations on agentic AI. The organization explained this by feedback on the draft, growing interest among financial institutions in using the technology, and its specific risks and control mechanisms.

The FSB also removed a recommendation to consider AI as “synthetic workers.” Instead, financial institutions were advised to take into account that AI agents do not think like humans and may unexpectedly interpret predefined goals.