Automation Anywhere is strengthening its push into agentic AI with a new set of AI-native solutions developed in collaboration with OpenAI, as enterprises look to move beyond task automation towards more autonomous, outcome-driven operations.
The latest capabilities combine Automation Anywhere’s Process Reasoning Engine (PRE) with OpenAI’s advanced reasoning models. The integration is aimed at closing the gap between AI-led reasoning and real-world enterprise execution, a challenge many organisations face as they attempt to scale AI from pilots into production.
At a technical level, OpenAI’s models are used for interpretation and reasoning, while PRE governs execution across enterprise systems. Together, they create what the companies describe as a reasoning-to-action loop, designed to allow AI agents to determine next steps and securely orchestrate work across applications.
“Traditional solutions automate work by following rigid steps, much like humans would, which often makes them brittle when things change,” said Mihir Shukla, CEO and Chairman of Automation Anywhere. “Our agentic solutions are fundamentally different; they’re designed for how AI agents work, enabling them to autonomously reason, solve problems, and adapt to changes — delivering exponentially greater ROI. With our Process Reasoning Engine and OpenAI’s reasoning models, we’re building agentic solutions that offer capabilities truly unlike anything in the market – it’s a true game changer.”
The move reflects a broader shift in enterprise automation strategies. While robotic process automation (RPA) helped organisations automate repetitive tasks over the past decade, many companies are now exploring agentic AI to handle more complex, variable workflows that require judgment, context, and adaptation.
Automation Anywhere said the new solutions are intended to provide enterprises with a faster path to agentic operations, offering pre-built and production-ready capabilities that can be deployed in weeks rather than months. The focus areas include high-impact functions such as finance, HR, IT operations, and customer service.
“Automation Anywhere is showing what’s next, AI that goes beyond automating tasks to redefining how work actually happens,” said Giancarlo “GC” Lionetti, Chief Commercial Officer at OpenAI. “Together, we’re embedding intelligence directly into core workflows so enterprises can move faster, work smarter, and drive meaningful outcomes.”
A recurring challenge with agentic AI in enterprises has been striking the right balance between autonomy and control. Fully autonomous systems can raise governance and compliance concerns, while overly constrained systems limit value. Automation Anywhere says its approach blends AI reasoning with deterministic execution and human-in-the-loop oversight to address this issue.
“Most agentic initiatives fail because they’re either too autonomous or too constrained,” said Dustin Snell, SVP of Agentic Solutions at Automation Anywhere. “What’s different here is that we deliberately blend agentic reasoning, deterministic execution, and human judgment into a single, governed flow. That balance is what turns promising AI experiments into reliable production outcomes for the enterprise.”
As enterprises continue to experiment with generative and agentic AI, the collaboration highlights how vendors are increasingly pairing large language and reasoning models with domain-specific orchestration layers to make AI usable, and governable, at scale.