South Korean banks and megachurches hit by AI-assisted cyberattacks

By Splitfeed EditorialLast updated Oct 7, 2026, 14:41 UTC

In early October 2026, hackers using a Chinese-developed AI tool carried out cyberattacks against several South Korean banks, with officials reporting that customer data from somewhere between 25,000 and 68,000 people was exposed. The lower figure of approximately 25,000 affected customers came from earlier official estimates reported on October 6, while South Korea's president cited the higher figure of up to 68,000 in subsequent remarks reported on October 7. Around the same period, South Korean megachurches opened separate probes into suspected AI-linked cyberattacks against their own institutions. South Korean authorities publicly identified the tool used as Chinese in origin and stated that it allowed attackers to carry out the intrusions without specialized technical skills.

The significance of these incidents extends beyond the data breach numbers. What makes the story consequential is the official attribution of the attacks to AI agents specifically, and the authorities' assessment that commercially available AI substantially lowered the technical barrier for carrying out financially motivated cybercrime. If that assessment is accurate, it marks a shift in the threat landscape, where sophisticated intrusion capabilities are no longer limited to state actors or highly skilled criminal groups. The dispute over the customer count, ranging from 25,000 to 68,000, also points to ongoing uncertainty in the official accounting, which matters for understanding the scale of harm and the regulatory response that may follow.

None of the three camps, Pro-AI, Anti-AI, or Middle Ground, had published reactions to this story at the time of writing. The Anti-AI camp would typically treat an incident like this as direct evidence that AI development poses concrete public safety risks and that insufficient regulation allowed dangerous tools to proliferate. The Pro-AI camp would likely argue that the problem lies with specific actors and misuse rather than with AI technology broadly, and might call for targeted enforcement rather than restrictions on development. The Middle Ground camp would generally be expected to focus on the need for security standards around AI tools, distinguishing between the technology itself and the policy failures that left institutions vulnerable.

The figures and framing to watch going forward include any final official count of affected customers, formal findings from the megachurch probes, and whether South Korean or international regulators move to classify or restrict the Chinese AI tool identified in the attacks. Any diplomatic response directed at China over the tool's origin would also be a significant development.