Salesforce Unveils Trusted Enterprise AI Harness Architecture
Salesforce has introduced its Trusted Enterprise AI Harness architecture to help enterprises securely build, manage, and scale AI. The architecture provides agents with the context, capabilities, and controls needed to understand business environments, make decisions, and take trusted action.
The Agentic Enterprise is changing how work gets done, with AI agents taking on increasingly complex responsibilities across organisations. They need to understand information, determine the right next steps, execute actions across systems, and collaborate with people and other agents. Salesforce's Enterprise AI Harness addresses this challenge by creating a trusted foundation around AI that enables agents to understand the business, reason and plan, take action, and operate within enterprise controls.
The architecture brings together six trusted capabilities: Trusted Context, Trusted Agency, Trusted Action, Trusted Governance, Trusted Security, and Trusted Models. These capabilities enable AI to access relevant information, understand customer and business context, determine appropriate actions, securely execute workflows, and operate according to enterprise policies and controls.