Hackers can weaponize hidden prompts revealed by downscaled images to trigger sensitive tool actions and achieve data exfiltration in Gemini CLI—and similar risks extend to Google Assistant and other production AI systems, according to new research by Trail of Bits. By exploiting how AI services routinely apply image scaling, the researchers showed that a benign-looking upload can morph […] The post Hackers Can Exploit Image Scaling in Gemini CLI, Google Assistant to Exfiltrate Sensitive Data appeared first on Cyber Security News.