Governments and businesses are facing a growing cycle of software vulnerability discovery and emergency repairs as artificial intelligence tools become more capable of scanning large volumes of code for security flaws. A November 21 analysis by The Washington Post describes how organisations are racing to repair weaknesses in ageing digital infrastructure while also defending against attackers.

AI Is Changing The Pace Of Vulnerability Discovery

The report says advanced AI tools can examine extensive codebases and identify flaws that may have remained hidden for years. That capability can help security teams find and fix weaknesses before criminals exploit them, but it also creates pressure to assess and patch a large number of systems in a short period.

Repair work is complicated by the interconnected nature of modern infrastructure. A change to one system can affect other services, and organisations must test patches carefully to avoid creating new failures. The Washington Post cited cybersecurity leaders who said the remediation effort could continue for 12 to 18 months.

The Same Technology Can Help Attackers

AI can also assist malicious actors by helping them scan for weaknesses, automate reconnaissance and produce convincing phishing material. The report described incidents involving attacks on South Korean banks and cyber operations that affected supermarket refrigeration systems, illustrating how digital threats can create consequences beyond computers and data.

The newspaper also reported that recent outages and emergency system changes were not necessarily part of one coordinated attack. Officials described several disruptions as consequences of rushed upgrades and efforts to fix vulnerabilities, rather than evidence of a single campaign by a foreign adversary.

Why Legacy Infrastructure Matters

Many essential services depend on software and hardware built over decades, with components maintained by small teams or open-source communities. These systems can be difficult to modernise because they support services that cannot simply be switched off for extended periods. The problem is especially serious where a failure in one supplier or network can cascade into other connected services.

The report’s central warning is that AI increases both the speed at which flaws can be found and the potential scale of attacks, while the work of safely repairing infrastructure remains slow. Organisations will need to combine automated security tools with careful testing, staff training, incident planning and sustained investment in resilient systems.

The accompanying image is a general computer-security photograph and does not depict a specific cyberattack or affected organisation. The full analysis is available from The Washington Post.