G7 plans drills to test responses to AI cyberattacks

G7 finance leaders have agreed to run simulated exercises to test their response to cyberattacks involving artificial intelligence, according to a Reuters report carried by The Indian Express on October 18. Japanese finance minister Satsuki Katayama described the agreement after a meeting in Bangkok.

The planned exercises will assess existing response procedures and help officials update their manuals, a Japanese finance ministry official told Reuters. Finance ministries and central banks will be involved, while private financial institutions could participate depending on the scenario being tested. The report does not give a start date.

A tabletop exercise is a structured discussion about how an organisation would handle a hypothetical emergency. The US Centers for Medicare & Medicaid Services’ information-security handbook describes participants working through their roles and responses without deploying equipment. This is preparation for a possible incident, rather than an announcement of a real attack.

Such an exercise can reveal whether a written plan is clear enough for the people expected to use it. It gives teams a setting to examine responsibilities and decisions before an emergency. The CMS guidance distinguishes this discussion format from exercises that actually test operational equipment and resources.

For the G7, the focus is how authorities coordinate when AI changes the cyber-risk picture. The reported agreement establishes a plan to rehearse that response; it does not establish that the drills have happened or that participating institutions have passed a new security test.