Using Microsoft Copilot in the Energy and Resources Sector

Frontline safety builds on the same foundation as Infrastructure and Operations.

• At the Start tier, safety managers can use Copilot Chat to summarise the sections of a safety plan relevant to the day’s tasks, extract the specific procedures and PPE requirements for a particular job, generate a tailored site checklist covering hazard controls and compliance points, draft a pre-job briefing for the crew, and format everything for distribution — a process that keeps safety documentation current and consistent without the administrative burden that often causes it to slip.
• At the Buy tier, health and safety inspectors can use Copilot to understand a new local regulation, update inspection checklists in Forms to reflect it, use Copilot’s multimodal capabilities to analyse photographs of workers near hazardous equipment and flag risks automatically, review a contractor’s project Health & Safety Plan, generate a draft safety violations report from checklist notes and site photographs, and draft an urgent communication to the head of H&S — all in the course of a single inspection.
• At the Extend tier, AI agents connected to the maintenance system enable assisted alarm investigation: the control room operator receives an alarm, asks the agent for a detailed description of the error code and the asset involved, identifies the most common causes from the O&M knowledge base, retrieves the resolution procedure from past incident records, generates a maintenance ticket summary if the alarm cannot be resolved remotely, and – once resolved – adds a new entry to the knowledge base so the next operator benefits from the experience.

• At the Start tier, operations planners can use Copilot Chat to analyse the week’s operational data for anomalies and emerging reliability patterns, identify the top risks across all sites and assets, rank them by impact and urgency, recommend mitigation actions with named owners, and produce a formatted summary for leadership, thereby transforming what is typically a reactive, backwards-looking reporting process into proactive risk management.
• At the Extend tier, energy traders gain significantly more capability: a Copilot Studio agent connected to trusted market sources collects current commodity insights; Copilot then generates a short-term market outlook with sentiment analysis; a further agent connected to regulatory databases conducts counterparty credit risk assessments drawing on five years of annual reports and mitigation strategies; Copilot in Word drafts trade settlement contracts from previous templates and relevant emails; and Copilot in Teams captures the contract review meeting and generates the list of required updates.

• At the Start tier, geoscience and subsurface teams can use Copilot Chat to scope a research question, surface the most relevant technical reports, seismic notes, and permitting documents, synthesise lessons learned and best practices from past projects, identify gaps in available data, and produce an annotated reading list for onboarding new team members, thereby compressing knowledge transfer that previously took weeks into a structured process that can be completed in hours.
• At the Extend tier, AI agents connected to permitting systems support the full environmental permitting workflow for new capital projects: identifying the relevant regulations for a given project type and location, comparing regulatory requirements across jurisdictions, generating the full structure of an environmental impact statement, populating sections with relevant historical data on hazards and site conditions, and enriching specific sections with supplier and equipment-specific information. For a sector where permitting delays routinely cost months and tens of millions of pounds in deferred revenue, this is one of the highest-value AI applications available.