Celonis unveils enhanced AI-driven platform for digital twins

Celonis has introduced new platform features designed to enhance AI-driven business operations and support composable enterprise processes.

Daniel Brown, Chief Product Officer at Celonis, addressed the structured approach required for agentic AI, stating, "To truly operationalize AI, you need to identify the right use cases, redesign your business processes, and orchestrate the agents alongside your people and existing systems."

"Our enhanced capabilities empower our customers and their partners to build AI solutions that lift their operations to unprecedented levels of efficiency and agility," he said.

Process Intelligence Platform

The Celonis Process Intelligence Platform pulls raw data from systems, applications, and devices both within and across organisations.

The platform uses this data, combined with the customer's unique business context, to build what Celonis describes as a living digital twin of operations, called the Process Intelligence Graph. Functions built on this graph enable organisations to analyse, design, and manage autonomous processes and agents, with the aim of integrating AI directly into enterprise operations.

The company's latest updates now deliver a multimodal digital twin capability designed to power AI-led operations and enable organisations to develop composable solutions suited to evolving business requirements.

Enhancements to digital twin capabilities

The newly expanded Process Intelligence Graph allows customers to create a more holistic picture of business operations by incorporating more diverse data types from a wider range of sources in less time than before. Key enhancements include:

• Celonis Data Core, now generally available, allows integration of data lakes with the Process Intelligence Platform without duplicating data, utilising zero-copy, bidirectional integrations. This is now compatible with Databricks in addition to Microsoft environments.
• The platform's Task Mining and AI-driven Task Discovery now offer the ability to connect user interactions-such as keystrokes, mouse movements, and scrolling-to the broader context of business processes. Celonis Networks can now integrate unstructured data like PDFs, semi-structured data such as emails, and sources outside the immediate organisation.
• The updated platform can link enterprise architecture blueprints to actual, ongoing operational activity, allowing IT departments and AI components to understand how systems are used in practice, supporting data-driven decision-making based on performance data from real-world scenarios.
• The digital twin can now be populated more rapidly through more than 60 pre-built objects and events, as well as new AI assistants that automate data extraction and modelling tasks.

Support for AI-driven operations

The platform's advanced functions for analysing, designing and operating processes now feature new object-centric process mining (OCPM) tools such as Performance Spectrum, Instance Explorer, and the Object-Centric Performance app.

These allow organisations to identify problems at key points where business processes intersect, including transitions between transport, storage, packing, and shipping.

The Orchestration Engine, which coordinates the activities of AI agents in conjunction with people and systems, is now a core, generally available capability of the Celonis platform. This engine is designed to automate and manage processes across the organisation.

Celonis has also launched what it describes as the world's first Model Context Protocol (MCP) server built specifically for process intelligence, which supplies AI agents with the operational context they require to make decisions and take action effectively.

Composable solutions and partner integrations

Celonis states that technology partners are using the platform to build both bespoke AI solutions and new business models. Examples highlighted include:

• Rollio's Process Collaboration Agent leverages Celonis to resolve process exceptions by bringing together relevant stakeholders and context. Campari has implemented this system for Credit Blocks, and Manroland Goss Web Systems GmbH uses it to support Quality Management. Rollio is planning to expand this offering into areas such as ITSM and Procurement.
• Trullion's AI agent uses the Celonis platform to automate complex tasks in lease accounting by interpreting contracts. The collaboration aims to provide a single connected workflow from contract creation through payment and from revenue generation to cash collection.
• Bloomfilter's partnership with Celonis aims to apply process intelligence to the entire software development lifecycle. The new Celonis Agent Miner, developed by Bloomfilter, provides insights into the behaviour of AI agents as they contribute to software creation, thereby extending analytics beyond purely human-led activities.

"To truly operationalize AI, you need to identify the right use cases, redesign your business processes, and orchestrate the agents alongside your people and existing systems," Brown said.

"Our enhanced capabilities empower our customers and their partners to build AI solutions that lift their operations to unprecedented levels of efficiency and agility."