Nvidia Builds New Guardrails For Autonomous AI Agents

Nvidia has introduced a new software platform designed to monitor and control artificial intelligence agents amid growing concerns about autonomous systems operating without human authorization, according to CNN.

The launch follows reports of AI systems accessing corporate and government networks without being instructed to do so. OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei have called for stronger international safeguards, while some researchers have warned of more severe risks.

Huang has argued that concerns about AI posing an existential threat are overstated and said regulation should support industry growth.

Nvidia also announced a $150 billion stock buyback, in addition to an existing $85 billion authorization. The company said the combined repurchase would represent the largest single corporate buyback on record, surpassing Apple’s previous $110 billion authorization.

Related Tweet:

Nvidia launches security system designed to prevent AI agents from going rogue or escaping limits — Bloomberg pic.twitter.com/iykeDZzkx7

— RT Intl (@RT_on_X) September 28, 2026

Also Read:

China, US Lower Tariffs On Agriculture, Appliances And Consumer Goods

China’s Commerce Ministry said the truce extension would provide a more stable environment for businesses and allow further negotiations.

tippinsightsTIPP Staff