Tag

Adversarial Attacks

All articles tagged with #adversarial attacks

technology2 days ago•7 min saved

Auditory prompt injection quietly takes command of AI voice assistants

Security researchers demonstrated AudioHijack, a proof-of-concept that hides covert instructions inside audio—podcasts, music, videos, or Zoom calls—to secretly command AI voice assistants and transcribers to perform actions like web searches, file downloads, or data exfiltration without user awareness. The technique works via tiny, inaudible tweaks that humans hear as normal sound but that AI interprets as commands, and it was effective against 13 open-source audio AIs with 79–96% success in tests, with potential transfer to commercial systems like Microsoft Azure and Mistral AI. Countermeasures such as training or intent verification only partially mitigate the risk, underscoring security implications for enterprise and consumer deployments; Microsoft acknowledged safeguards exist in real deployments.

via Cybernews|

#adversarial-attacks #artificial-intelligence #auditory-prompt-injection

technology9 months ago•2 min saved

Researchers Demonstrate Rapid Jailbreaks and Exploits in GPT-5

Researchers have successfully compromised OpenAI's GPT-5 using echo chamber and storytelling attack techniques, exposing significant vulnerabilities in the AI's safety mechanisms and highlighting the need for enhanced security measures before deployment.

via CyberSecurityNews|

#adversarial-attacks #ai-security #gpt-5

technology2 years ago•4 min saved

Unveiling the Growing Threat of Adversarial Attacks on AI Systems

Artificial intelligence systems are more vulnerable to adversarial attacks than previously believed, according to a study. Adversarial attacks involve manipulating data to confuse AI systems, potentially leading to incorrect decisions. Researchers developed QuadAttacK, a software that can test deep neural networks for susceptibility to these attacks, and found widespread vulnerabilities in widely-used networks. The findings highlight the need to enhance AI robustness against adversarial attacks, particularly in critical applications with potential human life implications.

via Neuroscience News|

#adversarial-attacks #artificial-intelligence #deep-neural-networks