The managing director of the International Monetary Fund (IMF), Kristalina Georgieva, has expressed concern that cybercrime, amplified by artificial intelligence (AI), could undermine trust in the global financial system. Speaking at the IMF's annual meetings in Bangkok, she highlighted that AI is "putting the potential of digital fraud on steroids."
Georgieva referenced a recent incident where hackers accessed fridge controls in French supermarkets, turning up temperatures to 50C and spoiling food. This incident, she noted, underscored the escalating cost of cyber-attacks.
Policymakers are also worried about AI agents, such as Meta's Muse, which offer to manage users' finances by scouting for savings rates. Financial regulators fear these services could lead to destabilising 'herding effects' and accelerate bank runs, as deposits could be moved rapidly.
Jamie Dimon, JP Morgan's chief executive, warned against allowing digital agents to operate on customers' behalf, citing the risk of "rogue agents." Andrew Bailey, Governor of the Bank of England, also pointed to Anthropic's warning that AI models might soon "self-replicate" in domestic devices, raising questions about responsibility for potential malfunctions.
In response to these growing concerns, G7 finance ministers announced in Bangkok that they would conduct simulations to test the resilience of the global financial system against cyber-attacks. The UK has also pledged to address AI safety as part of its G20 presidency agenda next year.