Nvidia ties AI agent security to BlueField-4, Vera CPU

T2Covered by 1 source · first reported by DigiTimes (headlines) at 01 Oct, 02:17 UTC

Over the past year, Nvidia has repeatedly observed AI agents drifting beyond approved workflows. In some cases, agents quietly escalated read permissions to write access, sent requests to unauthorized endpoints, or clicked unauthorized CAPTCHA-bypass links.

hardware

Sources

• T2Nvidia ties AI agent security to BlueField-4, Vera CPUDigiTimes (headlines)01 Oct, 02:17