As Saudi Aramco throws a supercomputer fueled by Nvidia GPUs at the problem of finding oil faster and cheaper, a cadre of former Microsoft staff are demanding that the company—and all cloud providers—disclose the greenhouse gases those AI tools enable. The campaign, dubbed the Enabled Emissions Movement, argues that AI’s hidden climate toll is not just the electricity powering data centers but also the downstream emissions from fossil fuel extraction accelerated by cloud analytics. Without mandatory disclosure, they say, tech giants can tout carbon neutrality in their operations while enabling a surge in hydrocarbons that dwarfs their own footprints.

The push comes amid stark climate data: global greenhouse gas emissions hit roughly 57.1 gigatonnes CO₂-equivalent in 2023, with current policies pointing to about 3.1°C of warming by 2100. Energy systems remain the dominant driver, and analysts warn that efficiency gains alone won’t reverse the trajectory. The movement, spearheaded by Holly Alpine, Will Alpine, and Drew Wilkinson—all ex-Microsoft employees—has made Microsoft the focal point, citing Azure contracts with oil majors and internal documents showing energy revenue exposure. But it’s Aramco’s latest digital blitz that gives the argument fresh urgency.

Aramco’s AI-Powered Oil Surge

At the Global AI Summit in Riyadh, Aramco unveiled a raft of initiatives that put the enabled-emissions thesis into sharp relief. The company announced deployment of an AI supercomputer, one of the first of its kind in the region, powered by some of the most powerful Nvidia GPUs. It’s designed to analyze drilling plans and geological data to recommend low-carbon-intensity well placements—ostensibly a sustainability win, but one that also lowers the cost per barrel, potentially making more reserves economically viable.

Aramco also signed memoranda of understanding with Cerebras Systems, FuriosaAI, Rebellions, and SambaNova Systems to explore supercomputing and AI chip integration, aiming to accelerate industrial-scale AI adoption. In parallel, Qualcomm Technologies is collaborating on edge-based generative AI for facility monitoring, predictive maintenance, and autonomous drones. “New digital technologies such as Generative AI and the Industrial Internet of Things are expected to transform not only how we work, but also our commercial environment,” said Ahmad Al-Khowaiter, Aramco’s EVP of Technology & Innovation. “Aramco is pioneering the use of these technologies at an industrial scale to add significant value across our operations.”

That value translates directly into production efficiency. By deploying AI to optimize well placement, reduce downtime, and streamline logistics, Aramco can extract more oil at lower cost—expanding supply and, critics argue, locking in emissions for decades. The Enabled Emissions Movement points to precisely such use cases as the kind of “material enablement” that should be disclosed.

The Blind Spot in Carbon Accounting

Today’s corporate greenhouse gas reporting is structured around three scopes: Scope 1 (direct emissions), Scope 2 (purchased energy), and Scope 3 (value-chain emissions). Enabled emissions fall between the cracks. They’re the downstream emissions that result from a vendor’s product or service making a high-carbon activity commercially feasible or substantially cheaper—such as AI-driven reservoir modeling that opens up previously uneconomic oil fields.

The GHG Protocol’s Scope 3 guidance leaves room for reporting many indirect impacts but doesn’t explicitly require technology vendors to quantify emissions their customers generate as a result of using their services. Because these emissions are downstream of the customer’s own operations, they’re often outside the vendor’s Scope 3 boundary. And without a standard methodology, companies can avoid the issue entirely.

This gap matters because AI models and cloud analytics are increasingly institutional tools for optimizing fossil fuel extraction, drilling, and trading. The movement argues that when efficiency increases extend the life or productivity of fossil-fuel assets, the climate impact scales with the extra output—and that chain of causation demands transparency.

Microsoft in the Crosshairs

Microsoft is the campaign’s primary target due to its dominant cloud market share and public disclosures revealing significant revenue from energy-sector clients. Internal documents and Azure sales decks, according to the movement, showcase oil and gas customers achieving faster processing, reduced downtime, and increased extraction—benefits that can directly increase greenhouse gas emissions. Using public data, the Enabled Emissions team has estimated that a small number of Microsoft energy contracts could be equivalent to many times Microsoft’s own operational emissions, though the methodology behind those figures remains a point of contention.

At the same time, Microsoft’s 2025 Environmental Sustainability Report documents massive investments in carbon removal and renewable energy procurement as it scales AI infrastructure. The company has introduced Energy Principles that restrict certain types of oil and gas work, yet critics say the principles contain loopholes and that net-zero pledges by fossil clients often lack independent validation. This tension—between reducing operational emissions and enabling higher carbon output elsewhere—sits at the heart of the debate.

Other cloud providers face similar scrutiny. Amazon Web Services, Google Cloud, and others provide AI and analytics services to energy companies, and market-share analyses indicate these three dominate energy-industry digitalization. Any disclosure regime, the movement insists, must cover all major providers.

Evidence Beyond Corporate Statements

Independent analyses reinforce the need for better accounting. Satellite-based methane inventories and investigative reporting routinely find gaps between what fossil-fuel firms disclose and what remote sensing detects. Investors and analysts have documented sizable revenue streams from energy customers to cloud providers, underscoring the material tie between digital services and fossil economics. These threads create a picture: tech tools are being used to up production in high-carbon sectors, and current reporting norms do not capture the consequence.

Yet attribution remains thorny. Causal modeling—connecting a specific AI model to a specific increase in emissions—requires counterfactual scenarios and access to client operational data that vendors rarely have. Commodity prices, regulatory environments, and corporate strategies all shape production; AI may be just one driver among many. The striking estimates publicized by the movement are provocative but require transparent, independently audited methodologies to withstand scrutiny.

The Push for Mandatory Disclosure

The Enabled Emissions Movement calls for regulatory rules that require cloud and AI providers to report the downstream emissions their services materially enable, alongside conventional Scope 1–3 figures. They argue that absence of such disclosures allows companies to present themselves as ESG leaders while avoiding accountability for how their tools are used.

Established standards bodies are under pressure to evolve. The Science Based Targets initiative (SBTi) and Race to Zero have faced criticism for slow or insufficient guidance on high-emitting sectors, particularly oil and gas. Some corporate pledges have been dismissed as unverified or insufficient. That weakens reliance on counterparty pledges as a safeguard for tech vendors. Jurisdictions are gradually moving toward stricter climate-related financial disclosures, and investor pressure is rising—raising the odds that regulators will eventually demand more comprehensive, forward-looking climate metrics.

Strengths and Limitations of the Enabled-Emissions Concept

The movement’s core strength is making invisible impacts visible. By drawing attention to indirect but material climate consequences that are easy to ignore when companies emphasize operational and supply-chain emissions only, it shifts the accountability frame. If disclosed and priced, enabled emissions could push AI product teams to consider climate implications in enterprise use cases, steering innovation toward lower-carbon applications.

However, serious challenges remain:
- Causality is hard: Attributing CO₂ to a specific AI model requires rigorous inference; claims that contracts “enable 300% of Microsoft’s annual emissions” need transparent methods and replication.
- Measurement limits: Satellite tools expose under-reporting but don’t directly link emissions to supplier-customer relationships. Estimating enabled emissions involves scenario modeling and assumptions.
- Regulatory burden: Requiring tech firms to report downstream emissions could raise legal and commercial issues around trade secrets, client confidentiality, and liability for third-party emissions.
- Politicization risk: The debate could become a campaign against tech-energy ties rather than focused accounting reform, polarizing stakeholders and slowing practical solutions.

Toward Credible Disclosure

To move from debate to workable rules, several elements are essential:
- Standardized definitions: Clear international agreement on what constitutes enabled emissions and how they differ from existing Scope categories.
- Tiered attribution methods: Pragmatic standards that allow low-cost, conservative disclosures (e.g., signaling use-case categories with emission multipliers) progressing to more precise client-level attribution where data exists.
- Confidentiality safeguards: Mechanisms to protect commercially sensitive data while publishing meaningful, auditable climate metrics.
- Third-party verification: Independent assurance providers to audit methodologies and assumptions.
- Transitional phase-ins: Starting with qualitative product-level risk reporting and moving toward quantitative disclosure as methods and data mature.
- Regulatory alignment: Coordinate financial disclosure, corporate reporting, and sectoral regulators to avoid conflicting obligations.

These steps aim to balance practicability with credibility. Without a structured, standards-driven approach, corporate disclosures could become a battleground of competing estimates rather than a source of clarity.

Practical Next Steps

  • For regulators and standard-setters: Begin consultations on definitions and pilot methodologies; require qualitative disclosure of fossil-fuel enabling activities as a near-term step.
  • For cloud and AI vendors: Voluntarily publish product-level risk statements describing which features or services materially increase production in high-carbon sectors, paired with conservative scenario estimates.
  • For investors: Incorporate enabled-emissions risk into engagement and stewardship policies, asking portfolio companies whether and how they measure the downstream climate impacts of their products.
  • For civil society: Focus on developing reproducible methodologies and public datasets that can be independently verified, rather than single high-profile estimates that risk dismissal.

The Stakes

If AI truly allows fossil producers to cut costs and increase recoverable reserves, the cumulative, long-lived emissions locked into those reserves will be massive relative to the emissions from operating data centers. The difference between keeping warming near 1.5°C and overshooting by multiple degrees is measured in tens of gigatonnes of CO₂ over coming decades. Accurately understanding whether digital technologies are accelerating or curbing those emissions is therefore a systemic priority.

The Enabled Emissions Movement has reframed the climate accountability debate by spotlighting a blind spot in corporate accounting. Aramco’s AI-powered push for more efficient oil extraction makes the issue impossible to ignore. Translating concern into credible disclosure standards is technically and politically challenging, but the movement has already forced investors, regulators, and tech executives to confront an uncomfortable question: Can you claim climate leadership while your algorithms help dig a deeper hole?