Salesforce is launching Agentforce 3, which includes enterprise AI agents with real-time monitoring, access controls, the ability to interoperate, and increased global manageability.

Salesforce has announced Agentforce 3, an important expansion of its AI agent software, which should improve scalability, security, and integration into the existing IT systems. At the center of this publication is the new Command Center, which provides real-time performance monitoring and the opportunity to integrate with third-party tools through the use of open standards.

"Agentforce 3 is responding to the skyrocketing demand of enterprises," Salesforce said, as it has witnessed a staggering 233% increase in AI agent adoption within the past six months, boasting more than 8,000 customers using the product. Among the first companies to adopt this technology were 1-800 Accountant and Engine, and they have already experienced enhanced resolution rates and workflow speed.

Executives Jayesh Govindarajan and Adam Evans emphasized the fact that the AI agents implemented by Salesforce have become a common part of its operations and not an experiment. The update also includes improvements in performance due to its use of faster AI models and ability to host the models of Anthropic Claude safely in the Salesforce infrastructure.

However, one of them is the use of Salesforce Agentforce by PepsiCo in a wider digital transformation through the introduction of AI agents to help it manage its globally distributed operations more efficiently. The platform integrates with other systems using an in-built protocol called Model Context Protocol (MCP) to facilitate connections across enterprise systems.

Currently available to most users, the Agentforce 3 product by Salesforce sets a new benchmark for AI agent platforms with its enterprise readiness.