Japan will establish a team to simulate cyberattacks using AI to identify weaknesses in technology systems, as organizations worldwide scramble to address critical vulnerabilities in PHP, a programming language widely used across the internet.
The announcement follows the disclosure of security flaws that triggered millions of hacking attempts this week, forcing governments and companies to temporarily shut down websites and suspend online services while they implement security measures.
Japanese Industry Minister Ryosei Akazawa said on Friday that organizations must adopt an attacker’s perspective to identify vulnerabilities before malicious actors exploit them.
‘An attack can succeed by getting through just one hole,’ Akazawa said, stressing the need for defenders to use AI proactively and stay ahead of emerging threats.
The disruptions have affected banking, airlines, railways, tax, and government services, highlighting the risks posed by vulnerabilities in software that supports much of the internet.
Japanese mobile carrier SoftBank said it was suspending its emergency alert system for mobile users while carrying out maintenance to address a PHP vulnerability.
During the suspension, customers will not receive early warnings about earthquakes, tsunamis, disasters or evacuation information through the affected system.
Other Japanese organizations have also taken precautionary measures. Brokerage Nomura, lender Mitsubishi UFJ Financial Group, airline ANA and East Japan Railway temporarily suspended some online services. Japan’s Ministry of Finance, National Tax Agency and Meteorological Agency also suspended certain online services.
On Friday, JR East said some group services had resumed, while others would return once confirmed safe. Nippon Paint said it had applied a security patch as a precaution and restored its website.
Also Read: Capcom Breach Exposes Data Linked to 11.77 Million CAPCOM ID Accounts
The vulnerabilities highlight AI’s growing role in both cyberattacks and security defense. Cybersecurity experts say attackers increasingly use AI to automate tasks such as scanning for software weaknesses and developing phishing campaigns, making cybercrime faster, cheaper, and harder to detect.
Eva Chen, chief executive of cybersecurity firm Trend Micro, said Japanese businesses could be targeted because of their extensive records. She said AI has also reduced the protection previously offered by language barriers, adding that Japan has become too comfortable with existing infrastructure.
PHP developers released fixes on Wednesday for vulnerabilities that had remained in the software for years before AI uncovered them. They urged users to ‘update immediately.’
The incident has prompted organizations to review their systems and address potential security gaps. Japan’s planned AI-powered cyberattack simulations aim to identify weaknesses before attackers exploit them, as governments and businesses face growing pressure to strengthen digital defenses.