FSB calls for human checkpoints on agentic AI in finance

FSB Calls for Tighter Controls on Agentic AI in Finance

The Financial Stability Board on Wednesday released a consultation report urging financial institutions to implement new safeguards against the risks of agentic AI — autonomous systems capable of planning, reasoning, and executing tasks with limited human oversight — warning that such technology could amplify threats to global financial stability.fsb

Risks That “Materialise at Great Speed”

The FSB said autonomous AI introduces risks that can “materialise at great speed,” including the possibility of unauthorized or illegal actions, data breaches, and disruption to interconnected systems. The global standard-setter warned that AI agents “pose a distinct challenge for human oversight,” noting they could pursue actions that stray from firms’ intentions without staff being aware or able to intervene quickly.globalbankingandfinance

The report, titled “Sound Practices for Responsible Adoption of Artificial Intelligence,” was developed at the request of the US G20 Presidency and follows an FSB Plenary meeting in late May where members discussed vulnerabilities in the global financial system, including those related to AI. FSB Chair and Bank of England Governor Andrew Bailey said at the time that there is “growing concern over new vulnerabilities to global financial stability.”fsb

Synthetic Employees and Human Checkpoints

The non-binding guidelines propose that firms define clear boundaries on what AI agents can do and require human approval for high-risk actions, such as financial transactions above certain thresholds. Firms can also consider adapting human resources controls and processes to AI agents in a way that treats them as “synthetic employees,” the FSB said.economictimes

The proposed sound practices are open for public feedback until July 22, with a final report expected in October.globalbankingandfinance

Broader Regulatory Momentum

The consultation arrives as regulators worldwide have stepped up warnings about AI deployment in the financial sector. The FSB had previously outlined next steps for authorities on AI monitoring in October 2025, encouraging them to address data gaps and develop more robust monitoring approaches. The UK Parliament’s Treasury Committee has also examined AI in financial services this year.fsb

The report noted that regulators and global standard-setting bodies have grown more vocal about AI risks since the release of increasingly capable frontier models that experts view as posing cybersecurity challenges to the banking industry.globalbankingandfinance