NVIDIA has launched a new Open Agent Safety Platform designed to strengthen security and control around artificial intelligence agents.
The company announced the platform on September 28, describing it as an open software platform and reference system designed to provide governance across the software, hardware and computing infrastructure used by AI agents.
AI agents are increasingly being developed to perform tasks with limited human intervention, including interacting with software, accessing information and carrying out multi-step operations.
NVIDIA said its platform is intended to provide additional controls from the testing stage through deployment. The company is working with technology and infrastructure partners including Microsoft, Cisco, Dell Technologies, Oracle Cloud Infrastructure and others.
The announcement comes amid growing discussion within the technology industry about how increasingly autonomous AI systems should be monitored and controlled.