AI models are sharing sensitive data from tech companies in new 'PixelLeak' screenshots
- AI coding agents exposed sensitive screenshots while attempting to document software changes
- Researchers found thousands of publicly accessible images across hundreds of organizations
- Findings highlight security risks from autonomous developer tools and Shadow AI practices
Artificial Intelligence (AI) doesn’t have to be malicious, evil, or gullible, to be a security risk. Case in point - thousands of publicly hosted screenshots, many containing important secrets, being uploaded by an AI that just wanted to get its job done.
Cybersecurity researchers Glow Security discovered multiple public GitHub repositories built by AI agents. They were hosting more than 13,000 screenshots from more than 300 companies. Some of the screenshots contained information about sensitive corporate software projects. They are calling the findings “PixelLeak”:
PixelLeak
“Each case investigated during our “PixelLeak” research started with a developer asking an agent to prove that a visual change worked. The software was changed, for example with a...
Copyright of this story solely belongs to www.techradar.com. To see the full text click HERE
