The transition from Robotic Process Automation (RPA) to agentic AI represents a significant leap for organizations, one that can initially appear daunting. However, SS&C Blue Prism is actively guiding its clients through this evolution, enabling a comfortable progression from established RPA frameworks to the more sophisticated capabilities of agentic automation.
This shift is not merely an incremental upgrade but a fundamental necessity driven by the escalating complexity of modern business workflows. Steven Colquitt, VP of Software Engineering at SS&C Blue Prism, highlights that traditional RPA, while powerful, was designed for a less dynamic operational landscape. Today’s environments are characterized by unstructured data flowing from diverse sources, mirroring the unpredictability of real-world interactions. “Inputs can vary, outcomes can shift, and decisions depend on context in real-time,” Colquitt explains.
Brian Halpin, Managing Director of Automation at SS&C Blue Prism, illustrates this with a practical example: processing a credit agreement. This might involve extracting and interpreting not just data points, but rather 30 to 40 distinct “answers” that require a degree of reasoning, a capability inherent in large language models (LLMs).
The concept of a “journey” is central to SS&C Blue Prism’s approach. Halpin elaborates, “We’re now saying we’re giving an AI agent the outcome that we want, but we’re not giving it the instructions on how to complete. We’re not saying, ‘follow step one, two, three, four, five.’ We’re saying, ‘I want this loan reviewed’ or ‘I want this customer onboarded.’”
Halpin foresees this outcome-oriented approach as the future trajectory for the market. However, he candidly addresses the current readiness, or lack thereof. “Is it ready for that? No. Why? Because there’s trust, there’s regulations, there’s auditability, stability, and security. We know LLMs are prone to hallucinations, they can drift, and if you change the underlying model, things change and responses get different.”
The path to fully autonomous, agentic workflows driven by non-deterministic AI is still under development, requiring substantial learning and adaptation. “There’s an awful lot of learning to happen before I think companies go fully autonomous and real agentic workflows are driven from that sort of non-deterministic perspective,” Halpin notes. He adds that the landscape is perpetually evolving, with new models emerging, reinforcing the notion of continuous progress.
SS&C Blue Prism serves thousands of clients who have already implemented robust automation processes, from established Centers of Excellence (CoEs) to operational digital workers. The company aims to help these organizations transition into the “world of AI,” often by integrating previously disparate functionalities.
“It’s been interesting,” Halpin observes. “As I talk to our customers, I see a common thread among companies right now where, in a lot of cases, AI has been established as a separate unit in a company. You go over to the process automation team, and they’re maybe not even allowed to use the AI.”
The challenge, as Halpin sees it, is to empower these teams with AI capabilities, blending them with existing process efficiencies to unlock further automation potential, aiming for that crucial additional 20-30% improvement in end-to-end processes.
To facilitate this, SS&C Blue Prism is set to introduce new technology designed to help organizations build and embed AI agents within their workflows, alongside robust orchestration capabilities. Attendees at TechEx Global, who participated in the Intelligent Automation conference, received in-depth insights into SS&C Blue Prism’s strategic direction and upcoming innovations.
Halpin emphasizes SS&C’s own extensive adoption of automation, stating, “[SS&C Technologies] are one of the biggest users of RPA in the world. We have over three and a half thousand digital workers deployed across the SS&C estate. We’re saving hundreds of millions in run-rate benefit. We’ve about 35 AI agents in production attached to those digital workers doing complex tasks, and really, we just want to share that journey.”
Original article, Author: Samuel Thompson. If you wish to reprint this article, please indicate the source:https://aicnbc.com/18564.html
