SLB and NVIDIA expand AI collaboration to industrialise energy-sector data infrastructure

Global energy technology company SLB (NYSE: SLB) announced on March 25, 2026 an expanded technology collaboration with NVIDIA (NASDAQ: NVDA) aimed at deploying AI infrastructure and domain-specific AI models across the energy industry at enterprise scale. The partnership spans three work streams: modular data center design, the development of a purpose-built AI Factory for Energy, and the optimisation of AI workloads across SLB digital platforms using NVIDIA accelerated computing. For SLB, this is a significant reorientation of its Digital segment toward infrastructure-as-a-product rather than pure software services. For NVIDIA, it deepens penetration into the energy vertical at a moment when the company is actively working to extend its GPU and AI platform dominance beyond hyperscalers.

SLB shares were trading near $50.51 as of March 24, 2026, within a 52-week range of $31.11 to $52.45. The stock has recovered substantially from its April 2025 lows, reflecting the broader re-rating of energy services companies as oil infrastructure spending cycles have held firmer than many expected amid geopolitical pressure. Morningstar carries a fair value estimate of $79.00, suggesting meaningful upside relative to current prices, though that estimate was set in January and predates more recent macro turbulence linked to Middle East tensions and oil price volatility. NVIDIA closed at $178.68 on March 25, 2026, up roughly 2.77% on the session, trading within a 52-week range of $86.62 to $212.19. The stock remains approximately 16% below its all-time high, range-bound in the $170 to $198 corridor as investors weigh the company’s $1 trillion Blackwell and Vera Rubin revenue ambitions against real-world macro headwinds.

Why is SLB building modular AI data centres for the energy industry in 2026?

The centrepiece of the operational announcement is SLB’s appointment as the modular design partner for NVIDIA DSX AI factories. Under this arrangement, SLB will manufacture data centre components offsite at its 3.1 million square-foot technology centre in Louisiana, which will then be assembled and deployed at energy company locations. The modular construction model is not new in data centre design, but its application to the energy sector carries specific advantages. Oil and gas operators, particularly those with remote production sites, have historically struggled with the lead times, labour constraints, and cost unpredictability that come with conventional data centre builds. Modular off-site manufacturing addresses each of these pain points simultaneously.

The strategic significance extends beyond construction logistics. By establishing SLB as the physical infrastructure layer for NVIDIA-powered AI deployment in energy, the partnership creates a bundled value proposition that competes directly with hyperscaler cloud offerings. Energy companies have been under pressure to move AI workloads off public cloud infrastructure for latency, sovereignty, and cost reasons, particularly for real-time operational applications. SLB and NVIDIA are effectively building the on-premise and edge alternative, scaled through a modular design that allows rapid capacity expansion without the overhead of bespoke engineering projects.

Execution risk here is real. SLB’s manufacturing history is strong, but pivoting a facility built for oilfield equipment and instrumentation toward precision data centre component production requires different quality systems, supply chain relationships, and technical workforce competencies. The Louisiana site will need to evolve, and that transition carries both capital and time-to-market risk. The company has not disclosed investment quantum for this initiative, which makes capital allocation discipline harder to assess from the outside.

What is the AI Factory for Energy and how does it differ from existing SLB digital platform offerings?

The second work stream, the AI Factory for Energy, is perhaps the most strategically ambitious element of the announcement. SLB and NVIDIA are developing what they describe as a reference environment: a pre-configured, domain-specific AI deployment stack running NVIDIA Omniverse libraries and NVIDIA Nemotron open models, integrated with SLB’s Delfi digital platform and Lumi data and AI platform. The intent is to give energy operators a deployable AI environment that is already calibrated for energy data types, without requiring each customer to build their own model training and inference infrastructure from scratch.

This matters because the energy industry’s AI challenge is not primarily a compute problem. It is a data organisation, domain model relevance, and integration problem. Energy companies generate enormous volumes of subsurface, production, and infrastructure data, but that data is structurally heterogeneous, historically siloed by asset or geography, and often only partially digitised. Off-the-shelf large language models trained on general internet data are of limited direct utility for seismic interpretation, well log analysis, or predictive maintenance of compressor fleets. Domain-specific models trained on curated energy datasets are considerably more valuable, but developing them requires resources and expertise that most operators do not maintain in-house.

SLB is positioning Delfi and Lumi as the data and AI orchestration layers through which those domain models are deployed and managed. The AI Factory for Energy, if executed well, would allow SLB to commoditise the data infrastructure challenge while capturing recurring platform revenue. The risk is that the reference environment becomes a lowest-common-denominator product that satisfies no single customer’s specific operational requirements particularly well. Reference architectures in enterprise software have a long history of being admired in vendor presentations and quietly abandoned in production.

How does the SLB and NVIDIA partnership alter the competitive dynamics in energy digital services?

The partnership creates a credible threat to several categories of competitor simultaneously. At the infrastructure layer, it challenges hyperscalers including Amazon Web Services, Microsoft Azure, and Google Cloud, which have been competing for energy sector AI workloads through their cloud platforms. SLB and NVIDIA are explicitly building the on-premise and modular edge alternative that allows operators to retain data sovereignty and reduce latency without dependence on public cloud. While hyperscalers have the advantage of scale, global presence, and established enterprise relationships, they lack the domain expertise and energy-specific data assets that SLB brings to this partnership.

At the software and analytics layer, the partnership intensifies pressure on specialist energy software vendors such as Halliburton’s iEnergy platform, Baker Hughes’ Leucipa, and a range of independent geoscience software and digital services companies. SLB has consistently invested in Delfi and Lumi as integrated digital platforms, and the NVIDIA collaboration adds a hardware and AI model layer that independent software vendors cannot easily replicate. The barrier to a competitor assembling an equivalent stack, encompassing domain expertise, proprietary training data, a validated modular hardware product, and a deep GPU partnership, is now considerably higher.

For NVIDIA, the SLB partnership represents sector penetration at a moment when the company’s go-to-market strategy has visibly shifted from selling GPUs to hyperscalers toward building sector-specific AI factory ecosystems. The energy vertical is attractive because it combines compute-intensive workloads, large proprietary datasets, and enterprise clients with substantial capital expenditure budgets. The March 23, 2026 announcement of the NVIDIA and Emerald AI initiative to use AI factories as grid assets further illustrates this multi-sector AI factory strategy, of which the SLB collaboration is the energy industry-specific expression.

What does agentic AI deployment in energy operations mean for operational efficiency and workforce requirements?

The announcement specifically references industrial-scale agentic AI as a component of the AI Factory for Energy. Agentic AI, systems capable of autonomous multi-step decision-making and action rather than passive response to queries, represents the most consequential and least proven dimension of this collaboration. In energy operations, the prospective applications range from autonomous well optimisation and predictive maintenance scheduling to supply chain management and regulatory compliance monitoring. If these applications mature to production deployment, the implications for workforce structure within energy companies are significant.

The energy industry has historically absorbed technology-driven efficiency improvements through workforce rationalisation rather than workforce growth. Automation of field operations, remote monitoring, and digital twins have already reduced headcounts across many upstream operators. Agentic AI applied to decision-making workflows that currently require experienced engineers and geoscientists would accelerate this trajectory. For SLB, which employs approximately 109,000 people globally and provides engineering services to operators, the long-term strategic implication is a potential reduction in the labour-intensity of its own service delivery model, improving margin but also transforming what operators are actually paying SLB to do.

How does this SLB and NVIDIA expansion build on the 2024 Delfi and Lumi AI collaboration, and what has changed?

The partnership formally dates to 2008, when NVIDIA accelerated computing was first applied to SLB’s seismic imaging and subsurface visualisation software. The more recent 2024 announcement focused on generative AI solutions integrated with Delfi and Lumi. The March 2026 expansion moves the collaboration from AI experimentation and software integration into physical infrastructure deployment and industrial-scale production. The shift from software collaboration to hardware manufacturing partnership is strategically significant. It indicates that SLB’s leadership believes the energy industry’s AI bottleneck has moved from model availability to deployment infrastructure, and that ownership of that infrastructure layer creates durable competitive advantage.

The timing is deliberate. SLB’s Q4 2025 earnings, reported in January 2026, saw revenue of $9.74 billion against a consensus estimate of $9.55 billion, with earnings per share of $0.78 against an estimate of $0.74. That operational outperformance gives the company financial credibility to pursue capital-intensive strategic expansions. However, Morningstar’s commentary on the stock notes that the oil market cycle has turned negative amid oversupply and secondary impacts from US tariffs, meaning SLB’s customer base is likely to face capital expenditure pressure heading into 2026. A strategic pivot toward infrastructure and AI platform revenue offers SLB a partial hedge against cyclical oilfield services demand.

What are the regulatory, geopolitical, and data sovereignty risks in deploying AI infrastructure for global energy operators?

The deployment of AI infrastructure for energy operators is not a purely technical exercise. Energy companies operate in jurisdictions with varying attitudes toward data localisation, AI governance, and foreign technology dependency. National oil companies in particular, which represent some of the most AI-ready and compute-hungry operators in the world, increasingly operate under mandates to keep operational data within national borders and to prefer technology partnerships that include knowledge transfer and local employment components. SLB’s modular data centre approach has an inherent advantage here: physical infrastructure can be deployed locally and data need not leave the jurisdiction.

NVIDIA faces its own regulatory headwinds. US lawmakers have been actively examining export controls on NVIDIA chips, with reports in late March 2026 of legislative efforts to suspend NVIDIA’s licence to export chips to China. While the SLB collaboration is focused on western and allied energy markets, the broader regulatory environment around semiconductor exports creates supply chain uncertainty for any deployment that relies on NVIDIA hardware at scale. SLB and NVIDIA will need to build supply chain resilience into the modular data centre programme to ensure that geopolitical disruption to chip availability does not translate into delivery delays for energy company customers.

Key takeaways: What the SLB and NVIDIA AI factory deal means for energy, technology, and infrastructure investors

• SLB is repositioning its Digital segment from software services toward physical AI infrastructure, acting as modular design partner for NVIDIA DSX AI factories through its Louisiana manufacturing facility.
• The AI Factory for Energy reference environment combines NVIDIA Omniverse and Nemotron models with SLB Delfi and Lumi platforms, targeting the domain-specific AI deployment gap that generic cloud AI cannot address.
• The partnership creates a bundled on-premise alternative to hyperscaler cloud offerings for energy AI workloads, directly challenging Amazon Web Services, Microsoft Azure, and Google Cloud in the sector.
• SLB shares near $50.51 carry a Morningstar fair value estimate of $79.00, though that upside calculation was set before the current oil cycle downturn and tariff-driven capex caution became more entrenched.
• NVIDIA’s SLB partnership is part of a broader sector-specific AI factory strategy, alongside the Emerald AI grid initiative, reflecting a deliberate shift from GPU commodity sales toward vertically integrated AI deployment ecosystems.
• Agentic AI deployment in energy decision-making, if it reaches production scale, carries long-term implications for the labour intensity of both SLB’s service model and its energy company customers’ engineering workforces.
• Execution risk centres on SLB’s ability to transition its Louisiana manufacturing facility toward precision data centre component production, a meaningfully different operational challenge from its traditional oilfield equipment manufacturing.
• Regulatory risk around US semiconductor export controls, particularly for NVIDIA chips, could create supply chain disruption that affects the modular AI data centre deployment programme.
• Specialist energy digital services vendors including Halliburton’s iEnergy and Baker Hughes’ Leucipa face increased competitive pressure from a stack that now bundles infrastructure, domain models, and data platforms under a single partnership.
• The oil market cycle turning negative on oversupply and tariff impacts means SLB’s energy customers may constrain AI infrastructure spending, potentially slowing adoption of the very platforms SLB is now investing to build.