UiPath, a workflow automation platform, has unveiled several new GenAI features in its platform, including context grounding, new large language models (LLMs), connectors, and an extension for its “Autopilot” solution for developers and testers.

Context Grounding will enable users to improve the accuracy of GenAI models via retrieval augmented generation. Users can improve the quality of generated outputs by providing information from company-specific datasets, such as knowledge bases or internal policies.

The company’s new LLMs, DocPATH and CommPATH, are specifically designed for document and communications processing to enable the models to understand documents and a large number of message types. Moreover, the company’s new connectors will enable users to leverage a number of foundation models from IBM's watsonx.ai platform. Additionally, UiPath’s new Autopilot features will allow users to create automation, codes, and expressions with natural language. Developers will also be able to generate step-by-step test cases from requirements and any other source documents, automate test steps, and surface insights from test results.

Analyst QuickTake: UiPath has been steadily ramping up its GenAI capabilities and solutions since it launched its GenAI-powered automation tool, Autopilot, in October of last year. The company has since entered a partnership with Google Cloud to tap into Google’s Vertex AI service and enable users to infuse GenAI insights into their automation. It also launched an integration with Amazon Bedrock to allow users to integrate GenAI directly into UiPath Studio and Studio Web automation.