EU DMA Shifts Cloud Battle: Google Drops Complaint as Regulators Probe Gatekeepers

Google’s decision to withdraw its formal EU antitrust complaint against Microsoft’s cloud practices is a swift, strategic pivot that hands the argument over to Brussels’ newly launched Digital Markets Act (DMA) market investigations — a move that reshapes the fight over cloud portability, vendor lock‑in and the future regulatory architecture for hyperscale infrastructure.

Background / Overview​

The complaint Google Cloud filed with the European Commission in late 2024 accused Microsoft of commercial and technical practices that, Google argued, made it materially harder and more expensive for customers to run Microsoft workloads on rival cloud platforms. Those allegations centered on differential licensing, migration friction, proprietary control‑plane dependencies, and self‑preferencing through bundled managed services — issues that directly bear on the ability of enterprises to adopt multi‑cloud strategies. On 18 November 2025 the European Commission announced a trio of DMA‑era market investigations focused on cloud computing: two company‑level probes to assess whether Amazon Web Services (AWS) and Microsoft Azure should be designated as DMA “gatekeepers,” and a horizontal sectoral study to evaluate whether DMA remedies are fit for cloud infrastructure. The Commission explicitly identified obstacles such as interoperability limits, conditioned access to business user data, tying and bundling, and imbalanced contractual terms as focal points for evidence gathering. Days after that Commission announcement, Google formally withdrew its standalone complaint. Google framed the withdrawal as pragmatic: the company said it would redirect its evidentiary and advocacy efforts into the Commission’s DMA process, while continuing to press for openness and customer choice across the EU and beyond. Giorgia Abeltino, Google Cloud Europe’s head of government affairs and public policy, described the move in an updated Google Cloud post as withdrawing “in light of the recent announcement that the EC will assess problematic practices affecting the cloud sector under a separate process.”

---

Why this procedural shift matters​

From private complaint to public regulatory forum​

The change in forum — from a private Article 102 antitrust complaint to a Commission‑led DMA market investigation — is far from cosmetic. The DMA provides regulators with powerful, ex‑ante tools that differ structurally from traditional antitrust remedies. Rather than seeking a case‑specific remedy after a protracted finding of abuse, the DMA offers a menu of prescriptive obligations regulators can impose quickly on firms designated as gatekeepers. Those obligations can include:

• Non‑discrimination and anti‑self‑preferencing rules.
• Mandatory interoperability and API exposure.
• Enhanced portability and data‑access requirements.
• Significant fines and periodic penalty payments for non‑compliance.

Because the Commission can compel documents and testimony across multiple firms and stakeholders, Brussels’ process can produce a systemic picture of commercial practices across the cloud sector — and do so on a faster timetable than a single antitrust investigation. That capability is exactly what motivated Google’s tactical withdrawal: the DMA process can order broad remedies that benefit many customers and suppliers, rather than a narrow fix limited to one complainant.

Scale, speed and reach​

Practically speaking, a DMA designation or DMA‑derived remedy would reshape product roadmaps, commercial contracts and technical interfaces across the hyperscalers. Gatekeeper obligations are ongoing and structural; they require continuous compliance and reporting. For businesses and governments that depend on cloud infrastructure for critical workloads and AI scaling, those kinds of binding, pre‑emptive rules are consequential. The Commission has signalled an aggressive fact‑finding horizon: company‑level market investigations are expected to run on roughly a 12‑month timetable, with the horizontal sectoral review taking longer. That creates an unusually tight window for evidence, stakeholder input and bargaining over technical standards.

---

The underlying facts: market concentration and market shares​

A key reason Brussels is treating cloud infrastructure as a DMA priority is the current market structure. Multiple independent industry trackers show the hyperscalers — AWS, Microsoft Azure and Google Cloud — account for a substantial, and growing, portion of global cloud infrastructure spending.

• Synergy Research Group reports that the “big three” captured approximately 63% of enterprise cloud infrastructure spending in Q3 2025, with AWS around 29–30%, Microsoft around 20%, and Google roughly 13%. These snapshots are widely cited by industry press and regulatory briefings.
• Canalys and other market research firms report broadly similar distributions for 2025 quarters, with AWS holding the lead, Microsoft in second, and Google in third, though the exact percentages vary by dataset and timeframe. Analysts emphasise that small percentage shifts reflect large sums of spending given the scale of the market.

These figures matter because market power in cloud markets is not only a function of market share by revenue; it also depends on breadth of services, geographic footprint, specialised AI infrastructure, and the degree to which customers rely on provider‑specific managed services and integrated stacks. Those qualitative attributes underpin Brussels’ concern that certain technical or commercial features may reinforce dominant positions over time.

---

What Google actually alleged — and what remains to be verified​

Google’s original complaint and related public statements focused on a set of commercial and engineering levers that can generate switching friction. Summarised, the main allegations were:

• Differential licensing and surcharges: Microsoft’s licensing terms for Windows Server, SQL Server and other enterprise software may impose higher effective costs when those products run on non‑Azure infrastructure, which can disincentivise migration.
• Migration and egress friction: High or complex data‑egress costs, limited export tooling or contractual exit terms that increase the time and cost to move production workloads off Azure.
• Proprietary control‑plane and managed service dependencies: Managed services or control‑plane APIs that are tightly integrated in ways that make functional parity difficult to achieve on rival clouds.
• Self‑preferencing and marketplace mechanics: Product packaging and marketplace placement that advantage Microsoft’s first‑party offerings compared with independent vendors.

Many of these points are structurally plausible and are precisely the lines of inquiry the Commission is now pursuing. But several of the most dramatic numeric claims that circulated in earlier coverage — for example, references to license mark‑ups “up to 400%” or industry estimates of hundreds of millions in annual additional costs — rest on contract excerpts, selective procurement scenarios or proprietary vendor calculations that are difficult to verify in the public domain. Regulators will need to examine contract samples, billing data and migration test evidence to substantiate any such headline figures. Treat those sensational numbers as allegations pending documentary verification.

---

How the key players are likely to respond​

Google​

By withdrawing the complaint, Google is not conceding the substance of its arguments; rather, it is repositioning itself as a participant in a broader regulatory process where the Commission has the authority to craft systemic remedies. This move reduces the optics of a bilateral, commercially motivated suit and channels attention toward Brussels’ rulemaking power. Google can still submit evidence, lobby for specific requirements (e.g., mandated API exposure, portability standards), and push for remedies that would lower engineering costs for customers and partners wishing to run Microsoft workloads off Azure.

Microsoft​

Microsoft has consistently denied systemic misconduct and argued that its commercial practices are legitimate and oriented around customer choice, security and compatibility. In regulatory filings and public comments, Microsoft warns that heavy‑handed rules could harm innovation or raise costs for customers. Microsoft’s defense will likely centre on demonstrating commercial justifications for licensing structures, technical reasons for non‑portable managed services, and the practical constraints of standardising portability across complex enterprise stacks. Expect Microsoft to provide technical briefings and contract samples aimed at showing the commercial rationales behind its licensing and managed‑service design choices.

Amazon Web Services (AWS)​

AWS faces parallel scrutiny as a company‑level subject of the Commission’s probes. While AWS is not the subject of Google’s withdrawn complaint, the horizontal nature of the DMA review means AWS will be asked to explain pricing, portability and coupling practices in the context of the industry’s dominant players. AWS’s responses will likely emphasise competitive pricing, breadth of services, and investments in interoperability tools where they exist.

---

Practical implications for enterprises and public sector customers​

The regulatory tug‑of‑war has immediate, actionable implications for IT architects, procurement teams and compliance officers. Firms should treat the Commission’s market investigations as a potential structural shift that could produce enforceable obligations — or conversely, as a source of transitional uncertainty if remedies prompt vendors to change product packaging or pricing.
Key pragmatic steps:

• Review current licensing and contractual exposure for Microsoft‑stack workloads, including any clauses that condition updates, transfers, or support on specific hosting arrangements.
• Negotiate explicit portability and exit terms in new cloud contracts: test migration processes in non‑production environments and document technical and commercial frictions.
• Maintain a multi‑cloud design posture where feasible: containerise workloads, decouple managed services where possible, and treat portability as an operational requirement rather than an afterthought.
• Engage procurement, legal and architecture teams to map potential regulatory outcomes into contract escape clauses, pricing re‑opener terms, and governance triggers.
• Monitor the Commission’s calls for evidence and consider submitting anonymised billing or migration test data if an organisation is materially affected — regulatory evidence matters in these investigations.

---

Strengths of the Commission’s approach — and real risks​

Strengths​

• Broad investigatory reach: The DMA process permits systemic discovery across multiple market participants, which is better suited to uncover patterns that individual antitrust cases struggle to reveal.
• Ex‑ante remedies: DMA obligations can prevent harm before it becomes entrenched, potentially making it easier for rivals and customers to switch providers or interoperate.
• Policy coherence for AI resilience: Cloud concentration is not only a competition matter; it touches on AI scaling, resilience and sovereignty — considerations the DMA’s toolbox can address in tandem with industrial policy objectives.

Risks and trade‑offs​

• Technical feasibility: Mandating interoperability, API exposure or managed‑service parity is technically challenging. Overbroad remedies could force brittle integrations, increase attack surfaces, or impede feature innovation.
• Investment disincentives: Hyperscalers argue that forced disclosures or stringent non‑discrimination rules could reduce incentives for heavy infrastructure investment (specialised accelerators, region expansion) if commercial returns become constrained.
• Regulatory precision vs. blunt instruments: The Commission must balance ambition with technical specificity — poorly drafted obligations risk being gamed or causing unintended consequences for reliability and security.

These trade‑offs mean regulators must pair legal mandates with clear technical standards, staged pilots and input from standards bodies and the engineering community. The regulatory gains of increased contestability must be weighed against practical costs for enterprise operations and provider innovation.

---

What Brussels will need from the evidence phase​

The Commission’s investigators will seek auditable, granular evidence to decide whether gatekeeper designation is warranted and what remedies are proportionate. Expect requests for:

• Contract samples showing license terms and any differential pricing or surcharges tied to hosting choices.
• Billing datasets and rebate/discount schedules that show effective price differentials.
• Technical documentation and migration test results demonstrating how tightly integrated managed services or control‑plane APIs impose re‑engineering costs.
• Information about marketplace dynamics and whether first‑party offerings receive preferential placement or bundling advantages.

Regulators will need to triangulate across sources — vendor filings, customer testimony, independent hosters, and technical experts — to build a robust factual record. That requirement is one reason Google’s withdrawal makes sense tactically: the DMA process can compel precisely the kinds of cross‑market evidence Google cannot easily obtain as a private complainant.

---

Likely scenarios and timelines​

• Short term (next 3–6 months): Commission compiles documentary evidence and seeks stakeholder submissions; hyperscalers and large customers engage intensively with investigators. Companies update procurement terms and begin migration tests in earnest.
• Medium term (~12 months): Company‑level DMA market investigations produce findings. If Microsoft or AWS are found to act as important gateways, the Commission may designate gatekeepers or propose tailored obligations for cloud services. The horizontal study could recommend DMA adaptations or delegated acts specific to cloud markets.
• Long term (12–24 months and beyond): Any designation and remedy package would prompt compliance programs, product adjustments, and potential litigation in EU courts. The net effect could be more standardised portability mechanisms, but also new compliance burdens and operational trade‑offs.

These timelines reflect the Commission’s stated ambition for a roughly 12‑month company‑level fact‑finding process, though the horizontal review will likely extend further. The path to enforceable remedies is not guaranteed; the Commission may decide that some issues are better addressed through sector‑specific guidance or voluntary standards rather than formal gatekeeper designations.

---

Areas requiring cautious skepticism​

Several high‑profile claims in public debate should be treated carefully until regulators publish verified evidence:

• Any headline number claiming uniform license mark‑ups (e.g., “up to 400%”) across customer classes is likely overstated; such figures usually arise from specific contract examples and do not reflect a market‑wide average.
• Aggregate estimates of “cloud tax” damages in the hundreds of millions or billions of euros must be validated against representative billing datasets and contract mixes.
• Technical claims about degraded security or delayed updates outside a provider’s native ecosystem require operational evidence and engineering comparators to be credible.

Regulatory fact‑finding is the only reliable path to verify these claims, because much of the relevant documentation is commercial and confidential. Organisations should therefore avoid treating sensational figures as established facts and instead use them as a prompt to gather their own migration evidence for submission to the Commission.

---

What to watch next​

• The Commission’s public calls for evidence and stakeholder consultations in the coming weeks; these documents will frame the investigative agenda and deadlines.
• Any formal submissions by Google, Microsoft, AWS, trade associations, national regulators, and independent hosters — these will reveal the contours of the factual record.
• Early technical workshops or standardisation initiatives convened by the Commission, industry groups or standards bodies to explore pragmatic interoperability solutions.
• Market moves: new portability tooling, revised licensing offerings from Microsoft, or product changes from other providers aimed at preempting regulatory fixes.

Businesses should track Commission filings closely and be ready to submit anonymised or redacted evidence if their contracts or operations are affected.

---

Conclusion​

Google’s withdrawal of its EU antitrust complaint against Microsoft is more than a procedural tidy‑up — it is a recognition that the Digital Markets Act and Brussels’ market investigations represent a more potent, comprehensive venue for addressing the systemic portability and competition issues that surround modern cloud markets. That shift elevates the contest from a private complaint into a public policy debate with immediate ramifications for procurement, architecture and technical standards across the enterprise stack.
The Commission’s DMA‑era toolkit offers the potential to deliver meaningful remedies — non‑discrimination, interoperability and portability — but turning legal ambition into practical, secure, innovation‑friendly rules is technically demanding. Regulators must fashion targeted, technically realistic obligations while avoiding blunt mandates that erode reliability or incentives for infrastructure investment.
For enterprises, the prudent posture is proactive: document licensing exposure, run migration tests, negotiate portability clauses, and engage with the Commission’s fact‑finding where appropriate. The months ahead will determine whether Brussels’ approach leads to more contestable cloud markets — or whether the complexity of cloud infrastructure requires a more calibrated blend of regulation, standards and industry cooperation to preserve both competition and capacity for innovation.

---

Source: Bangalore Mirror https://bangaloremirror.indiatimes....about-microsoft/amp_articleshow/125649638.cms