Researchers develop a device that converts electrical signals generated by light–matter interactions into true random numbers to authenticate images SEOUL, South Korea, July 13, 2026 /PRNewswire/ — Developing secure hardware capable of generating truly random numbers is becoming increasingly important as artificial intelligence-generated forgeries and cyber threats grow more sophisticated. Now, researchers have developed a light-driven true random number generator and demonstrated how these numbers can be embedded into images as hidden security signatures to detect tampering. The device could be integrated into image sensors, enabling hardware-based detection of image modifications. While security systems based on pseudo-random number generators (PRNGs) […]

Adversaries can embed executable instructions into images and audio so multimodal models read hidden directives from pixels and waveforms, bypassing text-only sanitization and leaving no visible logs. These techniques—typographic (FigStep), steganographic, semantic, and audio methods like WhisperInject—transfer across models, achieve high success rates in tests, and can be executed in the physical world. #FigStep #WhisperInject

Find out how AI image protections were cracked. Learn the future of image security in the AI era.