TotalEnergies and Cognite Launch Three-Year Global AI Rollout for Upstream Operations

TotalEnergies expands Industrial AI with Cognite across global E&P operations

TotalEnergies has signed a three-year program with Cognite to deploy the company's data and AI platform across all operated Exploration and Production assets worldwide. The rollout spans the full value chain, from drilling to production, with clear targets: efficiency, reliability, and sustainability. For operations teams, this points to a unified data foundation and faster decision cycles across sites.

The move builds on years of work between the two companies to make large volumes of industrial data AI-ready. By streamlining access to contextualized data, TotalEnergies expects quicker adoption of digital applications, stronger safety and performance monitoring, and improved field-level decisions. The platform is expected to deliver faster analysis, dynamic asset visualization, and wider use of AI tools across the upstream portfolio.

"This partnership with Cognite marks a new milestone in our digital transformation," said Namita Shah, President of OneTech at TotalEnergies. "By creating the data foundation which unifies our industrial data globally and makes it AI-ready, we are creating the conditions to accelerate AI-driven solutions that will significantly enhance the safety, operational and environmental performance of TotalEnergies." Cognite CEO Girish Rishi stated the collaboration is built on a shared vision to scale Industrial AI and equip teams to unlock insights across global assets.

What this means for Operations

• Integrated data layer across drilling, completions, production, maintenance, and HSE.
• Faster time-to-insight with standardized models and asset context (tags, equipment, process data, and work orders in one view).
• Dynamic visualization of assets to support surveillance, troubleshooting, and shift handovers.
• Broader use of AI for performance monitoring, abnormal condition detection, and emissions tracking.

Where value likely shows up first

• Drilling and well operations: offset-well learning, rate-of-penetration optimization, NPT reduction.
• Production surveillance: choke management, flow assurance alerts, deferment analysis.
• Reliability: predictive maintenance for rotating equipment and critical instruments.
• Turnarounds: scope challenge, integrated schedules, and readiness reviews informed by live data.
• HSE and sustainability: flare reduction, methane detection workflows, compliance reporting.

How to prepare your site (next 90 days)

• Inventory critical data sources (historians, SCADA/DCS, CMMS, LIMS, well and drilling data) and confirm access paths.
• Prioritize 3-5 high-impact use cases with clear owners and measurable KPIs.
• Map key tags to equipment and work orders to improve context and lineage.
• Set up data quality checks (freshness, completeness, outliers) and define escalation rules.
• Align OT/IT on connectivity, security, and change management for model deployments.

Governance and change management essentials

• Data ownership: name accountable owners for systems, tags, and models.
• Cybersecurity: enforce least-privilege access and network segmentation for OT/IT interfaces.
• Model lifecycle: versioning, validation, and rollback plans tied to Management of Change.
• Human-in-the-loop: clear playbooks for alarms, recommendations, and override rationale.
• Training and adoption: short, role-based sessions for control room, maintenance, and production engineers.

Related moves in TotalEnergies' stack

This program follows the company's two-year agreement with Emerson's Aspen Technology in July 2025 to deploy the AspenTech Inmation platform across industrial sites. Together, these steps point to a broader digital architecture where contextualized data and AI apps can scale across assets. Operations teams should confirm integration points and data ownership to avoid duplicate pipelines and conflicting sources.

KPIs to track

• Unplanned deferment and uptime by asset.
• MTBF and MTTR on critical equipment.
• Time-to-root-cause from alarm to action.
• Data readiness time for new use cases.
• Emissions intensity and flare volumes.
• Leading safety indicators (e.g., high-potential incident precursors).

For an overview of Cognite's industrial data approach, see the company's platform page: Cognite. For context on plant-wide data infrastructure, explore AspenTech Inmation.

If your team is building practical AI skills for operations and reliability roles, explore focused resources here: AI courses by job.