SAS has enhanced its Viya platform by introducing new AI assistants and advanced agentic AI tools, including a Copilot feature, a Model Context Protocol Server, and an Agentic AI Accelerator.
These enhancements are aimed at business and analytics teams seeking to transition AI initiatives from isolated experiments to integrated applications within their existing workflows.
The SAS Viya Copilot consists of a suite of AI assistants that are embedded throughout the analytics life cycle. These tools facilitate data analysis, model creation, and decision support directly within the platform, negating the need for a separate chat interface.
According to SAS, the Copilot tools enhance a variety of tasks within Viya applications. These tasks include Q&A assistance for data discovery, model pipeline development, model management, decision intelligence, and environment management. Additionally, they provide AI-generated SAS and Python code, documentation, model guidance, conversational dashboard creation, and AI-assisted visual exploration.
SAS is also launching a Model Context Protocol Server, built on the open MCP standard. This server allows organizations to expose SAS analytics, models, and decision-making tools to external AI agents without the need to reconstruct the foundational logic in another system.
Moreover, SAS has introduced the SAS Agentic AI Accelerator, which is a framework for building, governing, and deploying AI agents within Viya. This package includes code, components, interfaces, and working practices suitable for teams that leverage no-code, low-code, and developer-led methodologies.
Existing Viya users can access these agent-focused tools via GitHub. SAS frames this development as a means to connect AI-driven insights, decisions, and actions across various business systems and teams.
Embedded Assistants
The Copilot feature is pivotal to this expansion. SAS stated that it integrates Microsoft Foundry, enabling users to engage with AI support within analytical workflows through natural language prompts in both coding and visual environments.
Rather than confining the offering to general assistance, SAS is also providing industry-specific versions. The initial versions include the SAS Asset and Liability Management Copilot for financial risk analysis and the SAS Health Clinical Data Discovery Copilot for clinical data exploration and analysis.
The financial services Copilot is designed to assist analysts in scenario setup, execution, and interpretation for risk management. In the healthcare realm, the clinical data tool aims to facilitate data exploration, cohort creation, and quality investigation through natural language queries.
SAS plans to expand these industry-specific Copilot capabilities to areas such as financial crime prevention in banking and optimization in manufacturing planning and supply chain processes.
Governance Focus
Governance plays a crucial role in this launch, especially as businesses consider how to utilize generative and agentic AI in production environments. SAS asserts that combining assistants, agents, and internal controls can empower organizations to maintain human oversight while automating a greater portion of their workflows.
Jared Peterson, Senior Vice President of Global Engineering at SAS, emphasized this perspective during the launch remarks.
“The role of human expertise in operationalizing agentic AI is not diminished by automation; rather, it’s enhanced,” Peterson stated. “With SAS Viya, organizations can integrate copilots and agents with human judgment, reliable data, and corporate governance to ensure AI not only generates outputs but also fosters responsible decision-making.”
Another component of SAS’s governance strategy is the SAS Retrieval Agent Manager (RAM). This no-code product, based on retrieval-augmented generation, is crafted to transform unstructured data into contextually relevant responses.
Currently available as a standalone product, SAS plans to incorporate these functions into Viya, granting agents and assistants access to more contextual internal data when generating responses and recommendations.
Broader Platform Push
The launch is indicative of a broader trend among software suppliers to integrate AI assistants and autonomous agents as inherent components of analytical and business software. Vendors are increasingly striving to differentiate between generic consumer-oriented chatbots and tools that are firmly connected to specific data, regulations, and workflows within organizations.
For SAS, the emphasis is on ensuring that domain-specific analytics and governance are central to these systems. The Viya Copilot is designed to execute analytical tasks grounded in industry-specific data models, terminology, and practices, rather than relying primarily on the broad knowledge of a large language model.
These developments also highlight how open standards are beginning to influence the market. By adopting the MCP standard, SAS is aligning its platform with a growing movement to connect AI models and agents to external software tools, potentially simplifying customer integration of Viya functions with third-party interfaces.
SAS states that these new features aim to help organizations develop AI systems that function across various teams and processes, all while remaining governed by internal controls.
