Is this the new cyber normal?

AI can now generate exploits for newly disclosed vulnerabilities within hours, not weeks.

A series of severe cybersecurity vulnerabilities have emerged, causing widespread panic. The most significant is PHP2Shell, affecting an estimated 70% of the internet, including major organizations like Wikimedia Foundation, the UK government, and Japanese banking systems. This vulnerability, with a CVSS score of 9.2, was discovered using AI models like GPT-5.6 Sol and Gemini 4 Argon. Despite its discovery by security firms Calif and Wiz, and a prior audit in 2024 by the PHP Foundation, the exploit was rapidly disseminated and exploited in the wild before patches could be widely applied, due to alleged leaks from Calif and Chainguard. On the same day, eMudhra, an Indian certificate authority, announced its exit from the SSL/TLS certificate business after it was found to be using extremely weak private keys for its intermediate certificates, a flaw that a single AI prompt could reportedly exploit. Separately, cloud provider Railway disclosed a breach where hackers accessed and decrypted environment variables, potentially targeting companies involved in AI inference.

The rapid acceleration from vulnerability disclosure to widespread exploitation, driven by AI, signifies a fundamental shift in the cybersecurity landscape. This trend suggests that traditional patching timelines are becoming obsolete, forcing organizations to adopt more proactive and rapid response strategies. The proliferation of AI-generated "exploit slop" also means that discerning genuine threats from noise will require enhanced human expertise, potentially creating a bottleneck in security operations.

This new paradigm challenges the efficacy of current security measures and underscores the arms race between AI-powered offense and defense. Companies relying on older software or slower patching cycles face increased risk, and the demand for advanced AI-driven security solutions is likely to surge. The incident with eMudhra also highlights the critical importance of secure key management, even in seemingly robust infrastructure, and the potential for AI to rapidly expose such weaknesses.