Salesforce Inc. is stretching its data and artificial intelligence tentacles further into the public cloud, announcing expanded strategic integrations with both Amazon Web Services Inc. and Google Cloud.
The partnerships, announced at Dreamforce 2026 today in San Francisco, aim to break down enterprise data silos by bringing rich and valuable customer relationship management data, AI agents and workflows into its partners’ public cloud environments.
Salesforce’s work with AWS is focused on bringing AI agents into the place where enterprises get their day-to-day work done, while enabling broader model choices for users and enhanced voice capabilities. For instance, Salesforce business context and actions are being made available natively within AWS’ business-focused agent Amazon Quick via the Model Context Protocol. That means Amazon Quick will be able to access live CRM data from Salesforce and leverage it to perform autonomous work on behalf of sales and customer service teams.
In another move, Salesforce said a number of AWS’s frontier agents, beginning with the AWS DevOps Agent, are being integrated within Slack, its enterprise-focused messaging and collaboration tool. With AWS DevOps Agent in Slack, teams will be able to resolve infrastructure problems without having to leave that platform, the company said.
Meanwhile, Salesforce’s Agentforce platform is being given access to Amazon Bedrock’s library of foundation models, including those from Anthropic PBC, OpenAI Group PBC and Nvidia Corp., meaning users have more choice about how to power their Agentforce agents. In addition, Salesforce said it’s extending its Data 360 Zero Copy integration to AWS Glue-managed Apache Iceberg tables and the Amazon Aurora database, which makes it easier to ground AI agents in enterprise data.
Finally, Salesforce is enhancing real-time support for Agentforce users by adding Agent2Agent bidirectional audio support with Amazon Connect Customer over Websockets.
AWS Vice President of Data and AI GTM Rahul Pathak said enterprises need to build on their own data, with the applications and infrastructure they’re already using, to get the most value out of their AI investments. “These new integrations with Salesforce give customers data access without migration, model choice at the right economics and AI agents where their people already work,” he said.
A unified infrastructure for AI agents
Salesforce doesn’t have any preferences when it comes to public cloud infrastructure, for it used the occasion to announce a number of similar integrations with Google Cloud. The goal here is to build a “unified AI stack,” the company said.
The integration will enable cross-platform agent reasoning and actions that run on shared infrastructure, with Salesforce workloads brought onto Google Cloud via Hyperforce, which is a framework for running the company’s software on public cloud platforms. Salesforce said Hyperforce on Google Cloud is already supporting live customer traffic, and expects broader customer migrations to take place in the fourth quarter.
“What changes here is the infrastructure underneath our customers, not the trust they’ve built with us,” said Salesforce President and Chief Trust and Infrastructure Officer Meir Amiel. “Hyperforce running natively on Google Cloud will extend the same security, compliance, and resilience standards we hold ourselves to. Enterprises will get Google Cloud’s flexibility without giving up an inch of the trust bar they expect from Salesforce.”
There’s also a big integration with Google’s Gemini Enterprise platform to look forward to. Salesforce said customers will be able to surface their CRM data and other capabilities directly within Gemini Enterprise, without needing to do any integration work.
Instead, Gemini takes advantage of Salesforce’s new headless architecture, which is also built on MCP and brings Salesforce capabilities to any AI model or agent. That means Gemini can now be used to perform tasks like sales pipeline health checks, account summaries, and case triage across sales, service, finance and advertising workflows.
The Gemini integration also sees Tableau’s data analytics and visualization capabilities brought into Gemini. That will help ensure that every response is grounded in Salesforce data, meaning more accuracy and relevance, Salesforce said. And going in the other direction, Gemini is being brought into Agentforce, so Salesforce customers will also be able to develop agents powered by Google’s flagship AI model.
Finally, Salesforce said it’s integrating its Commerce Cloud platform with the Universal Commerce Protocol. That will make it possible for shoppers to discover and purchase products from directly within Google Search or Gemini, with payments processed via Salesforce’s secure infrastructure.
Kevin Ichhpurani, president of the Global Partner Ecosystem at Google Cloud, reinforced the need for business context to succeed in AI. “By bringing Salesforce into Gemini Enterprise, customers can now build trusted agents that reason across their business context and securely take action,” he said. “And with Hyperforce on Google Cloud, customers gain greater choice in how they deploy Salesforce on our AI-optimized infrastructure.”
