Agentforce
Last updated October 1, 2026
Agentforce is Salesforce's platform for building, running and monitoring AI agents inside Salesforce. An agent is software that can understand a request, plan the steps to handle it and then take actions, not just reply with text. Salesforce launched Agentforce in late 2024, and it replaced the earlier Einstein Copilot assistant.
An Agentforce agent has a few parts. Administrators give it a job description and plain-language instructions. They group its work into subagents for specific tasks, and give it a library of actions it may use, such as a Salesforce Flow, Apex code (Salesforce's programming language), a prompt template or a connection to an outside system. Ready-made actions and agent templates from Salesforce partners are sold on AgentExchange. Salesforce's reasoning engine breaks each request into steps and chooses which actions to run. The agent works from the company's own Salesforce records and connected data, and it can hand a conversation to a person when a request is outside its scope.
A note on the name. In October 2025, Salesforce launched Agentforce 360 as its broader platform. Its help pages also put "Agentforce" in front of product names, such as Agentforce Sales for Sales Cloud. As of October 2026, its product website uses the older names, such as Sales Cloud and Service Cloud. Agentforce remains the name of the AI agent platform itself.
Why it matters for lenders
Lenders that run on Salesforce can use agents for routine work: answering borrower questions about balances, drafting follow-up emails, collecting missing documents, or summarizing a file for a loan officer. Done well, that frees staff for judgment calls.
The risks are real, though. An agent that can take actions can take the wrong one. Lenders should limit which actions each agent may use, require a human in the loop for anything that affects credit or money, and log every output. In April 2026, the main US bank regulators replaced their model risk guidance, and the new version states that generative and agentic AI are outside its scope. That leaves lenders to set their own controls for now.
How it works in practice
A business lender builds a servicing agent in Agentforce. When a borrower writes in asking for a payoff amount, the agent finds the loan, runs the lender's existing payoff flow and drafts a reply with the quote attached. A servicing specialist reviews and sends it.
The agent is not allowed to change loan terms, waive fees or move money. If the borrower asks for a payment deferral, the agent creates a case for a person and tells the borrower when to expect an answer. Every draft and hand-off is saved on the loan record, so the servicing manager can review what the agent did each week.
Related terms
AI Agent (Agentic AI)
An AI agent is software that can work toward a goal on its own: it breaks a task into steps, uses tools such as databases, documents and other software, and takes actions, checking its progress as it goes. Agentic AI is the broader term for systems built this way. Unlike a chatbot, an agent can change data and trigger processes, so it needs firm limits.
Human-in-the-Loop
Human-in-the-loop means a person reviews and approves what an AI system prepared before it takes effect. The AI can gather data, draft a summary or suggest an action, but a named person decides and is accountable. In lending, it keeps credit decisions, adverse action notices and money movement under human control, with each review and override recorded.
AgentExchange
AgentExchange is Salesforce's marketplace for AI agents, apps and add-on capabilities that extend Salesforce and Slack. It started in March 2025 as a separate marketplace for Agentforce components such as agent actions and templates. By September 2026 Salesforce had folded AppExchange and the Slack Marketplace into it, so AgentExchange is now the single place to find, try and install Salesforce solutions.
Financial Services Cloud
Financial Services Cloud is Salesforce's industry version of its CRM for banks, credit unions, wealth managers and insurers. It adds financial data models, such as households, financial accounts and goals, plus industry workflows on top of Sales Cloud and Service Cloud. Salesforce's help pages call it Agentforce Financial Services, while its product website uses Financial Services Cloud as of October 2026.
Explainable AI
Explainable AI is AI whose outputs come with reasons a person can understand and check. It shows which factors drove a score, flag or recommendation, and the explanation must truly reflect how the system reached it. In lending, explainability lets a lender give accurate reasons for a credit denial, test for unfair treatment and answer examiners and funders.
See how lenders put this into practice
FUNDINGO runs origination, servicing and reporting on one loan record inside Salesforce.