On Thursday, 16 October 2026, in Bangkok, the capital of Thailand, the finance ministers and central bank governors of the Group of Seven (G7) advanced economies met on the sidelines of the International Monetary Fund's (IMF) annual meeting. At the end of the meeting, the group agreed to hold "tabletop" exercises against AI-enabled cyberattacks targeting banks and the financial system. Japan's finance minister Satsuki Katayama announced the agreement after the meeting, Reuters reports.
According to Katayama, the agreement reflects the need to coordinate responses to growing cybersecurity threats. Tabletop exercises make it possible to practice decision-making, information sharing, and coordinated action under pre-prepared scenarios — without affecting real systems: participants discuss a hypothetical situation and agree on their actions.
Where and When the Meeting Took Place
The meeting that produced the agreement was held in Bangkok as a separate session within the IMF's annual meeting. It brought together the finance ministers and central bank governors of the G7 advanced economies. Japan's finance minister Satsuki Katayama briefed the public on the outcomes: it was his Thursday statement that revealed the details of the exercise plan and confirmed its adoption at G7 level.
That the meeting took place on the sidelines of the IMF annual meeting reflects the fact that G7 finance chiefs were in Bangkok at the time — the separate session was arranged within the main event on the international financial agenda.
Sequence of Events
The agreement continued a chain of events that unfolded over several days. First, the IMF warned this week that an AI-enabled cyberattack on a financial institution's critical services could undermine trust. Then, on Thursday in Bangkok, G7 finance ministers and central bank governors met and agreed to hold tabletop exercises; the agreement was announced by Japan's finance minister Satsuki Katayama. On the same day, the Bank of England said it was working with banks on advanced cyber-recovery options — making it two major statements in one day on the financial sector's readiness for AI attacks.
How the Exercises Will Work
The exercise plan is based on bespoke simulations developed by G7 finance ministries. They will replicate AI-enabled cyberattack scenarios: participants will test how they manage an attack situation under existing procedures. The results will serve two purposes — first, to assess how effective existing protocols are in a real-world situation; second, to update the playbooks for countering such incidents. Feeding the results into the playbooks turns the exercises into a practical tool that directly improves defense documentation.
The finance ministries and central banks of the G7 countries will take part in the exercises. The fact that the simulations are developed specifically by G7 finance ministries ensures the scenarios account for the specifics of the financial system and response mechanisms. According to a Japanese finance ministry official speaking to journalists, private financial institutions may also be brought in depending on the nature of the simulations — in some scenarios the banks themselves will become exercise participants.
"The G7 finance chiefs agreed to hold tabletop exercises to guard against AI-enabled cyberattacks," Katayama said after the meeting in Bangkok (Reuters).
The IMF Warning
The G7 agreement was reached against the backdrop of a warning issued by the IMF this week. The Fund noted that an AI-enabled cyberattack on a financial institution's critical services could undermine trust — meaning the threat targets areas critical to the uninterrupted functioning of the financial system. The warning's emphasis on trust is no accident: confidence in financial services is one of the system's core pillars, so an attack on critical services could trigger a crisis of confidence alongside the technical problem.
This warning is not limited to G7 countries: an AI-enabled cyberattack could target the critical services of a financial institution in any country — Uzbekistan's banking system is no exception.
The Bank of England Is Acting Too
On Thursday, the same day as the G7 meeting, the Bank of England also announced that it was working with banks on advanced cyber-recovery options. A deputy governor's speech stressed that defense systems can no longer be assumed immune to AI-enabled cyberattacks — the focus is not only on preventing attacks, but also on plans to restore systems when they occur.
"They can no longer assume that their defense systems are immune to AI-enabled cyberattacks," the Bank of England deputy governor said on Thursday (Reuters).
