Summary
- HCLTech will modernise infrastructure and end-user IT operations across more than 200 M Group locations.
- The programme spans water, energy, communications, transport, defence, industrial, and public-sector activities.
- AI Force will sit inside a wider managed-services overhaul where reliability and operational integration matter more than standalone AI features.
M Group is moving more than 200 UK and Irish locations onto an AI-led managed-services model with HCLTech, putting automation into the technology estate behind businesses that maintain water, energy, communications, transport, and other essential infrastructure.
HCLTech has been selected to modernise infrastructure and end-user operations across the group, using its AI Force service-transformation platform. The contract reaches across M Group’s water, energy, technology and communications, transport, defence, industrial, and public-sector activities rather than a single office environment.
That breadth creates a different technology problem from an ordinary workplace-AI rollout. Field engineers, office employees, transport teams, and staff supporting defence or public contracts may all depend on different devices, connectivity, applications, and access rules, even when the organisation wants one service model underneath them.
The programme is intended to simplify that estate, improve service reliability, and create a more consistent digital foundation as M Group continues integrating businesses and capabilities.
A distributed estate changes automation
Managed-services providers have increasingly introduced AI into IT operations, where software can classify support requests, identify recurring faults, recommend fixes, automate routine changes, and correlate information across infrastructure and user environments.
Those applications are less conspicuous than employee-facing generative AI, but they can affect a larger share of the technology estate because they sit inside the machinery used to keep systems running.
In M Group’s case, the commercial result will depend less on the number of processes described as AI-enabled than on whether HCLTech can reduce incidents, shorten resolution times, and make services more consistent without creating new dependencies or brittle automation.
That threshold is higher in infrastructure businesses because IT failures can reach physical operations. Not every office-system outage interrupts an essential service, but organisations maintaining energy, water, communications, and transport assets have less room for technology changes that make field work harder during an incident.
Modernisation has to remove complexity
Technology estates rarely consolidate as quickly as corporate structures. Organisations often retain overlapping applications, support processes, identity systems, licences, and infrastructure inherited from earlier operating models.
AI can make that complexity easier to navigate without necessarily removing it. A support agent capable of guiding employees through several legacy applications may improve service while leaving the underlying cost and maintenance burden intact.
More durable savings emerge when automation arrives alongside rationalisation: common device policies, consolidated service management, fewer duplicated applications, clearer identity controls, and more consistent operational data.
A fragmented estate also creates more exceptions for automated systems to understand. Multi-site programmes therefore continue to depend on conventional disciplines including asset discovery, migration planning, governance, and change management even when AI operates part of the service layer.
Critical infrastructure raises the resilience threshold
The sectors served by M Group are also under growing pressure to understand technology dependencies and third-party risk. Moving more IT operations into a strategic supplier can create economies of scale and access to specialist skills while concentrating dependency at the same time.
Service levels, privileged access, disaster recovery, subcontractors, incident escalation, and data flows become part of the resilience design rather than contract details to be revisited later.
Automated actions also need enough observability to show what changed, why it changed, and which system triggered the action. Where a recommendation can affect employee access or production technology, approval boundaries need to remain clear.
The strongest evidence from the programme will consequently be operational rather than promotional: fewer interruptions, faster fixes, more consistent access, and less duplicated technology across a sprawling infrastructure organisation.










