Salesforce, Anthropic Expand Partnership For Trusted AI In Regulated Sectors

Salesforce and Anthropic have expanded their partnership to deliver trusted, enterprise-grade AI for regulated and data-sensitive industries, positioning Claude as a foundational model within the Agentforce 360 Platform. The collaboration focuses on scaling compliant adoption of advanced AI across sectors such as financial services, life sciences, healthcare, and cybersecurity — all within Salesforce’s secure trust boundary.

“Salesforce and Anthropic share a vision for a trusted AI ecosystem that puts customers at the centre,” said Marc Benioff, Chair and CEO at Salesforce. “By bringing Salesforce directly into Claude — and Claude’s intelligence into Salesforce and Slack — we’re giving every company the power to work in entirely new ways.”

Companies including CrowdStrike and RBC Wealth Management are early adopters, using Claude through Amazon Bedrock in Agentforce to automate client workflows and streamline operations without compromising compliance or data privacy.

Trusted AI For Regulated Environments

Anthropic is now the first LLM provider fully integrated within the Salesforce trust boundary, ensuring all Claude AI operations occur in a virtual private cloud under Salesforce’s compliance infrastructure. This provides a secure environment for institutions where data sovereignty, traceability, and regulatory adherence are non-negotiable.

As part of this initiative, Anthropic will collaborate with Salesforce on industry-specific AI use cases—starting with financial services—bringing the power of Claude to Agentforce Financial Services for intelligent portfolio management, regulatory monitoring, and client communication workflows.

Dario Amodei, CEO and co-founder of Anthropic, explained, “Regulated industries need frontier AI capabilities, but they also need the appropriate safeguards before they can deploy in sensitive systems. We’ve built Claude to deliver both: the performance and the safeguards.”

The partnership deepens the integration between Claude and Slack, leveraging Slack’s Model Context Protocol (MCP) to make AI assistance available directly within enterprise collaboration tools. Employees can summarize long threads, analyze shared documents, and access Salesforce CRM insights—all without leaving Slack.

Salesforce engineers are also using Claude Code, Anthropic’s AI-powered programming assistant, to enhance productivity across internal development workflows. The Claude Code integration within Slack enables developers to read code specifications, retrieve context, and generate documentation automatically, reducing repetitive work and shortening release cycles.

Building the AI Ecosystem for Regulated Industries

Anthropic’s Claude models hosted on Amazon Bedrock will further enhance Salesforce’s Agentforce 360 Platform, enabling AI agents that can reason over domain-specific datasets while ensuring that sensitive workloads never leave the secure Salesforce environment.

Customers in wealth management, insurance, and healthcare are using these tools to:

• Summarise client portfolios and policy documents.

• Flag compliance requirements and automate consent tracking.

• Generate actionable reports combining CRM and regulatory data.

According to Rohit Gupta, Head of Digital Advisor Platforms at RBC Wealth Management, “Because of Anthropic on Amazon Bedrock and Agentforce, we’re able to help our advisors with their most time-consuming task: meeting prep. This has saved significant time, allowing them to focus on what matters most – client relationships.”

The Agentic Enterprise Vision

This partnership reinforces Salesforce’s broader Agentic Enterprise vision revealed at Dreamforce 2025—a future where domain-specific AI agents can execute business logic autonomously while remaining transparent, compliant, and explainable.

By pairing Anthropic’s responsible AI design with Salesforce’s data governance and security, both companies aim to set a new standard for deploying trusted AI in sensitive sectors, balancing innovation with accountability.