Japan Logs More Cyber Incidents in Nine Months of 2026 Than in All of 2025

September marks a sharp rise

According to TrendAI™, September alone saw 86 incidents, up about 18% from August and 37% from July. The advisory followed a statement by Masaaki Taira, who chairs the Liberal Democratic Party's National Cybersecurity Strategy Headquarters. He said that "cyberspace is in a state of emergency."

The party's headquarters met on Oct. 9 in response to a series of cyberattacks on Japanese businesses and called on the government to strengthen public-private cooperation. Breaches at retailers, financial services firms, a car-sharing service and a rail operator have exposed data of millions of customers, according to the release.

The government's National Cybersecurity Office has sent guidance to ministries, local governments and private companies. It covers keeping security protection up to date, using strong passwords and tightening security across supply chains.

Web shells and the role of AI

The TrendAI™ Incident Response team has also seen more incidents in Japan. In recent engagements, attackers exploited known vulnerabilities in public-facing applications to plant web shells on web servers. That gives them a foothold from which they may attempt privilege escalation, lateral movement or access to connected databases. TrendAI™ says it has published Japanese-language analysis of these cases along with hardening guidance.

TrendAI™ states that it has not established a connection between the recent incidents in Japan and the use of AI by attackers. It does argue that AI lets attackers accelerate reconnaissance, analyze vulnerabilities and produce convincing phishing in languages they do not speak. In the company's view, this weakens the protection that languages such as Japanese once offered against foreign attackers. Automation also makes smaller organizations, including suppliers, more attractive targets.

Ryan Flores, Co-lead of Forward-Looking Threat Research at TrendAI™, said AI's ability to help attackers find targets and automate operations "makes cyberattacks scalable." Robert McArdle, Director of the same group, said "AI has zero cost for cybercriminals" and that the current surge "should be seen as the new baseline of attacks."

Recommended defenses

TrendAI™ lists six areas for organizations to address:

• Patch internet-facing systems first. Keep an inventory of public web servers, applications and cloud services, prioritize vulnerabilities known to be actively exploited, and use virtual patching while fixes are tested.
• Harden web servers. Restrict write access to application code, isolate file-upload directories, prevent uploaded content from running as scripts, and monitor for unauthorized changes.
• Watch database access. Monitor sensitive operations and abnormal access patterns, and run applications with minimum-privilege database accounts.
• Check suppliers. Ask hosting and cloud providers how they would notify you of an incident, review which third parties hold customer data, and confirm critical services can recover independently of a single provider.
• Strengthen identity verification. Review processes that rely solely on uploaded photographs of identity documents and adopt chip-based or other stronger verification.
• Bring AI usage into scope. Identify AI applications, models and agents in use, assess their provenance and reputation, and review their access to sensitive data and internal systems.

Vision One and upcoming AI security tools

The company also promoted its TrendAI Vision One™ platform, which it describes as built to address AI security risks from shadow AI, sensitive data sent to external services, and AI agents connected to internal systems with excessive permissions.

TrendAI says it is introducing new capabilities to the Vision One AI security suite in December 2026, including AI Gateway and AI Risk Insights solutions with digital asset reputation scoring. According to the company, they are meant to deliver visibility, observability, ownership and governance over AI use. TrendAI™ Vision One AI Risk Insights, including the AI digital asset reputation assessment capability, is scheduled for general availability in January 2027.

TrendAI™ says its researchers and incident responders are tracking the activity behind the current wave of incidents in Japan and will publish further findings as the analysis continues.

What it means for organizations

The advisory points to a practical message for Japanese companies and their suppliers: known vulnerabilities in exposed systems are the immediate entry point seen in recent cases, and smaller firms in supply chains are increasingly attractive targets. TrendAI™ frames the current volume not as a peak but as a baseline that will grow. Organizations that want to discuss their exposure can contact their TrendAI™ account team.