Securing AI systems against evasion, poisoning, and abuse
Adversaries can intentionally mislead or “poison” AI systems, causing them to malfunction, and developers have yet to find an infallible defense. Read more www.helpnetsecurity.com
Share this story, choose a platform
Brought to you by BridgeTower Media
Free Weekly Newsletter
Recommended content
Complete Communicator: Avoid having your assumptions become your convictions
Complete Communicator: Avoid having your assumptions become your convictions By Jay Sullivan As lawyers, we are taught to be discerning, [...]
Six telltale signs for detecting deepfakes
While AI programs like Microsoft's new VASA-1 are generating increasingly lifelike virtual characters, there are methods available to detect manipulated [...]
Cisco warns of large-scale attacks against VPN, SSH services
The "brute force" campaign detected in March attacks security equipment by Cisco, CheckPoint and others with a barrage of usernames [...]
How AI-enabled knowledge management can elevate small and midsize firms
An AI-aided knowledge management initiative can help smaller firms exploit their best work product and collective knowledge, and compete with [...]
