Nvidia releases more tools and guardrails to push companies to adopt AI agents

Nvidia is releasing three new NIM microservices, or small independent services that are part of larger applications, to help companies bring additional control and security measures to their AI agents.

One of these new NIM services aims at content security and works to prevent an AI agent from generating harmful or harmful outputs. Another works to keep conversations focused only on approved topics, while the third new service helps prevent an AI agent from jailbreak attempts, or remove software restrictions.

These three new NIM microservices are part of it Nvidia NeMo GuardrailsNvidia’s existing open source collection of software tools and microservices aimed at helping companies improve their AI applications.

“By applying several lightweight and specialized models such as guardrails, developers can cover the gaps that can occur when only more general global policies and protections exist – since a one-size-fits-all approach does not properly secure and control agent AI workflows complex, “u press release he said.

It seems that AI companies may be beginning to understand that companies adopting their AI agent technology will not be as simple as they initially thought. While people like Salesforce CEO Marc Benioff recently expected There will be more than a billion agents fleeing Salesforce in the next 12 months alone, the reality will probably look a little different.

A recent study by Deloitte expected that about 25% of companies are either already using AI agents or expect to in 2025. The report also predicted that by 2027 about half of companies will be using agents. This shows that, while companies are clearly interested in AI agents, they have not adopted AI technology at the same rate that innovation is happening in the AI ​​space.

Nvidia likely hopes that initiatives like this will make the adoption of AI agents seem safer and less experimental. Time will tell if this is actually true.