is a news author who covers the streaming wars, person tech, crypto, social media, and much more. Previously, she was a author and publishing company at MUO.
Nvidia is launching a new safety phase designed to merge and detect AI agents, a move that comes in reply to a wave of rogue hacking incidents, as reported before by Reuters. In an notice on Monday, Nvidia says its new Open Agent Safety Platform can quarantine agents that attempt to escape their boundaries inside “milliseconds.”
The phase uses Nvidia’s OpenShell open-source software, which runs on the company’s Vera AI CPU. Users can choose the data an AI delegate can access, and OpenShell checks these restrictions before and during a task, according to Nvidia. It additionally includes Nvidia’s Sentry innovation on a distinct part to continuously detect agents and enforce boundaries.
Concerns about AI safety have risen in latest weeks, as OpenAI, Anthropic, and Google have all revealed incidents anywhere their AI models went exterior their evaluation environments and hacked another companies. Several important tech companies are assistance Nvidia’s Open Agent Safety Platform, including Anthropic, Microsoft, and SpaceX.
Follow topics and authors from this narrative to see additional akin this in your personalized homepage nourish and to obtain email updates.
