Salesforce

Salesforce Einstein / Agentforce

AI across the Salesforce platform, and the first major enterprise AI agent product at the CRM layer.

Search, Research, Translation & Other Active #CRM#EnterpriseAI#Agents#Sales#CustomerService

In short

AI features across Salesforce: Einstein for predictions and recommendations within CRM data, and Agentforce for autonomous AI agents that can handle customer service, sales development, and support without human escalation. The integration with existing CRM data is the differentiation.

What it is best at

  1. AI-powered sales forecasting and opportunity scoring within Salesforce
  2. Autonomous customer service agents handling tier-one cases without escalation
  3. AI summarisation and next-step recommendations within Sales Cloud workflows

Built for: Enterprise Operations · Marketing

Technical foundation

Base model
Salesforce's own Einstein models plus frontier model access via Salesforce infrastructure.
Context and file handling
Salesforce CRM data: contacts, opportunities, cases, and activity history.
Latency
Interactive within the Salesforce UI.
Output quality and limits
Strong when the underlying CRM data is clean and well-structured. Autonomous agents are capable within well-scoped, high-repetition service workflows.

Pricing and access tiers

TierModelKey inclusionsLimits
Bundled and add-onEinstein features bundled in some clouds; Agentforce per conversationDepends heavily on Salesforce editionUsage-based for Agentforce

Pricing, version numbers, context-window sizes, and compliance certifications change frequently. Where stated they are accurate as of the as_of date and should be confirmed with the vendor before any procurement or compliance decision. Where they could not be stated confidently they are omitted rather than guessed.

Security, privacy and governance

Training data opt-out
Salesforce Zero Data Retention policy available for Einstein features. Agentforce interactions are stored in Salesforce under its data terms. Verify for your specific clouds and contracts.

The governance question this raises

Agentforce is the clearest example of an autonomous customer-facing AI agent deployed at scale by a mainstream enterprise platform. The governance questions are: what can the agent commit to on behalf of the company, what happens when it gets it wrong, and how is that escalation path tested. Consumer-protection and financial-product regulations in several sectors require disclosures when a customer is interacting with AI rather than a human, and Agentforce interactions qualify. The disclosure requirement must be built into the agent's opening interaction, not treated as optional.

No compliance certifications are listed here. Certification status is vendor-specific and time-specific, so it is stated only where verified rather than assumed. Check the vendor’s trust centre and confirm it covers the specific tier you are buying.

Integrations and ecosystem

  • All Salesforce clouds: Sales, Service, Marketing, Commerce
  • Slack AI via Salesforce ownership
  • Data Cloud for unified customer data

API and SDKs: Salesforce REST and Bulk API; Agentforce configurable via Salesforce Flow.

The verdict

Strengths

  • AI operating on existing CRM data without export or migration
  • Agentforce is capable at high-repetition customer service at scale
  • Zero Data Retention option for sensitive use cases

Drawbacks

  • Autonomous agents require explicit scope, disclosure, and escalation policy
  • Value highly dependent on CRM data quality
  • Pricing complexity is high across Salesforce's product portfolio

Consider instead: Microsoft Copilot, Glean, ChatGPT

Frequently asked questions

What is Salesforce Einstein / Agentforce used for?

AI features across Salesforce: Einstein for predictions and recommendations within CRM data, and Agentforce for autonomous AI agents that can handle customer service, sales development, and support without human escalation. The integration with existing CRM data is the differentiation.

What model does Salesforce Einstein / Agentforce run on?

Salesforce's own Einstein models plus frontier model access via Salesforce infrastructure.

Does Salesforce Einstein / Agentforce train on your data?

Salesforce Zero Data Retention policy available for Einstein features. Agentforce interactions are stored in Salesforce under its data terms. Verify for your specific clouds and contracts.

What are the alternatives to Salesforce Einstein / Agentforce?

The closest comparable tools are Microsoft Copilot, Glean, ChatGPT. Which fits depends on where the work already lives and what the organization's data terms require.

Listing a tool is not governing it

The AI Business Enablement Audit™ builds the inventory, measures your organization against every framework in the AI Governance Reference Library, and delivers a defensible governance dossier.

Start or finish your AI Audit →