ðŸ‡ðŸ‡· Croatian, 7 November 2026, AI-powered attacks
Microsoft Discovers How AI Accelerates Theft of Sensitive Data
Automated systems using AI agents can find exposed access keys in minutes, allowing attackers to access cloud services.
Microsoft researchers have discovered that attackers are using AI-powered tools to quickly find and exploit exposed access keys in cloud services. The tools, known as ARTEX, use AI agents to search public code repositories and find sensitive data, such as access keys and passwords. This allows attackers to access cloud services, including Microsoft’s own Azure and Microsoft 365, in a matter of minutes. The researchers found that attackers were using ARTEX to search for specific organizations, keywords related to access keys, and popular cloud services, in an effort to find exposed data that could be used to access protected systems. The researchers also found that attackers were targeting access keys for AI services, which could allow them to access sensitive data and disrupt cloud services. Microsoft is warning organizations to be vigilant and regularly review their code repositories, limit the permissions of their applications, and monitor unusual access to cloud services. They are also advising organizations to replace exposed access keys as soon as possible and check if they have been used without authorization. The researchers found that attackers were using a pattern of repeated attempts to access cloud services, with each attempt using a different version of the API or method of sending requests. This suggests that the attackers are using automated systems that can adapt to changing circumstances and reduce the need for human oversight during attacks. The researchers also found that attackers were using compromised identities to access additional sensitive data, which could allow them to access other systems and disrupt cloud services. Microsoft is urging organizations to be aware of the risks and take steps to protect themselves from these types of attacks.