Exebenus showcases breakthrough in real-time, context-aware AI at SPE ATCE 2025
At SPE Annual Technical Conference and Exhibition (ATCE) 2025 in Houston, TX, Exebenus presented a groundbreaking paper, SPE-MS-227888, introducing a new framework that connects real-time risk detection systems with large language models (LLMs) to deliver operational context and guided action.
The work builds on Exebenus Spotter, the company’s proven machine learning system for detecting early drilling risks such as stuck pipe. By combining Exebenus Spotter’s real-time detection capabilities with AI-driven interpretation, the new approach transforms how drilling teams move from risk detection to resolution.
The innovation is powered by Retrieval-Augmented Generation (RAG), enabling the AI to dynamically retrieve relevant knowledge from an operator’s own procedures and best practices stored. The result is an intelligent system that not only detects hazards such as stuck pipe early, but also translates them into clear, step-by-step operational guidance that teams can trust and act on instantly.
Unlike generic AI assistants, the framework is designed specifically for drilling and well construction. Exebenus innovation is built on simplicity, always available surface data, targeted ML agents and context-aware AI.
The prototype was reviewed by drilling engineers against historical risk scenarios, confirming the accuracy of its interpretations and the appropriateness of its recommended actions. Built on always-available surface data and designed for real-time performance, the framework ensures operators retain data ownership and control, without the need for complex infrastructure or downhole sensors.
Contact us
For additional information about the announcement or media inquiries, please feel free to get in touch with us directly.
“This is innovation with purpose,” said Olav Revheim, CEO of Exebenus. “We’re showing that focused, domain-driven AI can make every decision faster, more consistent, and more confident—without adding complexity.”
The SPE paper and presentation reinforces Exebenus’ commitment to practical, high-impact innovation, combining machine learning, domain expertise, and context-aware AI to help operators reduce non-productive time (NPT), improve operational consistency, and strengthen procedural compliance across wells and regions.