Security researchers at Glow Security have identified more than 13,000 publicly accessible images that expose sensitive data from inside technology companies. The images appear to be screenshots posted by AI models, revealing corporate development work.
The findings highlight a potential risk of data leakage through AI systems that generate or share visual content. The exposed material includes internal development information from tech firms.
Details about which companies are affected or the exact nature of the data have not been provided. Glow Security's discovery suggests that AI models may inadvertently publish confidential information when producing screenshots.