IBM and Salesforce have announced an enhanced strategic partnership that combines the capabilities of IBM's watsonx AI and Data Platform with the Salesforce Einstein 1 Platform. This collaboration aims to provide customers with greater flexibility and choice in AI and data deployment, allowing teams to make data-driven decisions and take actions seamlessly within their workflows.
Key aspects of this expanded partnership include bidirectional data integration, flexible large language models (LLMs), prebuilt actions and prompts for CRM solutions, and a commitment to responsible AI development.
"Clients need enterprise-grade foundation models that are trusted, performant, and cost-effective. Choosing the right foundation models is essential to every organization's AI strategy and a key determinant of how quickly projects move from pilot to production. IBM and Salesforce are working together to make it easier to navigate the complex generative AI landscape by helping clients select the right LLM for their business needs," said Ritika Gunnar, General Manager, Product Management, Data and AI, IBM.
Flexible large language models
Organizations using generative AI for CRM applications require LLMs tailored to their specific needs. Salesforce's "Bring Your Own Large Language Model" strategy enables customers to choose and integrate their preferred LLMs with the watsonx.ai platform and Salesforce Einstein 1 Studio. This integration includes access to IBM's Granite series AI models and the ability to use custom LLMs built on watsonx.
IBM watsonx supports enterprises in building, scaling, and governing custom AI solutions, offering a variety of AI models from different providers that can be customized with their own data. IBM's Granite models, trained on high-quality enterprise datasets, are designed to meet strict data governance standards and provide strong performance and developer-friendly features, along with intellectual property indemnification from IBM.
The integration of IBM Granite models with the Einstein 1 Platform allows joint customers to leverage these models for diverse purposes, such as industry-specific content generation, summarization, and classification. These models are optimized for trust, transparency, and scalability across enterprises. To facilitate quicker ROI, pre-configured prompt templates and copilot actions, powered by IBM data and services, will be available in Einstein 1 Studio and can be invoked in flow and apex.
Industry-specific prompts and actions
With IBM Consulting's industry expertise, IBM is developing industry-specific prompt templates and copilot actions that are expected to be available in Einstein 1 Studio later this year. These pre-built actions and prompts will be accessible across Salesforce applications, including Einstein Copilot, Salesforce's conversational AI assistant for CRM.
Initial use cases focus on the public sector, such as government contact centers, eligibility determination, and licensing and permitting. These templates are designed to work with both foundation and Granite models to enhance engagement in sectors like automotive, energy, utilities, financial services, and public sector.
IBM Consulting is assisting joint customers in establishing sustainable and responsible AI frameworks to scale AI-driven Salesforce CRM initiatives, accelerating time to value while mitigating risks.
Bidirectional data integration
IBM has joined the Salesforce Zero Copy Partner Network, allowing secure, bidirectional zero copy integrations with Salesforce Data Cloud. This enables customers to access IBM data via watsonx.data from within Salesforce Data Cloud without moving data, simplifying security management and reducing risks.
Zero-copy integration enables businesses to maintain their data in its original location while accessing external data from Salesforce Data Cloud, creating a comprehensive customer view. This approach minimizes the need for custom integrations and complex data pipelines, maximizing technology investments and ensuring historical data remains accessible for analytics and Generative AI.
IBM Granite is anticipated to be available in Salesforce early next year. For more details on IBM Granite, please visit the provided page.
"Salesforce and IBM are well positioned to make it easier for customers to get the most value from their data — no matter where it resides — and establish a foundation for customer engagement and trusted AI. With the addition of bidirectional data integration, companies can harmonize all of their data faster, fueling actionable insights that empower teams to make data-driven decisions and deliver amazing integrated experiences across all touchpoints and channels for every customer interaction," said Rahul Auradkar, EVP & GM, Unified Data Services & Einstein, Salesforce.
