AI-Driven Vulnerability Surge Forces Microsoft to Set New Patch Tuesday Records
Microsoft engineers have been working overtime to keep up with the rapid pace of software vulnerabilities discovered by AI models. In June, Microsoft set a new record for patch Tuesdays, fixing around 200 flaws in Windows and other products. The following month, July's patch Tuesday was even bigger, with over 570 security holes patched.
According to sources familiar with Microsoft's security work, the recent surge in vulnerabilities is due to the use of AI models like Anthropic's Mythos and OpenAI's cybersecurity-focused model. These models are able to quickly identify potential weaknesses in software, but they also create a high volume of patches that need to be verified and applied.
Microsoft's latest patch Tuesday is expected to set another record, with over 650 security fixes for Windows alone. This represents six times the usual number of patches issued before the AI models arrived. Engineers have been busy verifying hundreds of important patches, which include remote code vulnerabilities and privilege escalations.