Who would feel the impact of conduct that global antitrust authorities are investigating?
When competition law enforcers in Washington, London, and Brussels talk about “lock‑in” agreements, they are describing the quiet power of contracts and licensing terms that can tilt a market. In cloud computing, that tilt can be expensive and harmful, and no one faces risks greater than California companies. Mid‑sized brands, fintechs, SaaS platforms, health‑tech innovators and many others in the state increasingly depend on how easily they can leave a cloud, not just enter one. But, do contractual restrictions stand in their way? Complaints to authorities suggest they do.
According to the Data Center Map, California stands out as one of the nation’s primary cloud‑infrastructure hubs. It hosts 326 data centers with concentrations in Santa Clara, Los Angeles, San Jose, and San Francisco. This extensive physical infrastructure is complemented by a mature statewide cloud‑adoption strategy. The California Cloud Services program is aligned with the federal Cloud Smart framework and emphasizes standardized cloud architecture, modernized security practices, and workforce upskilling.
Contracts that raise rivals’ costs and make customers’ exits painful are squarely in the antitrust frame.
A distant third in this space, Google filed an antitrust complaint with the European Commission in 2024, alleging that Microsoft’s cloud licensing practices unfairly lock customers into Azure through steep markups and contractual barriers for running Windows workloads on rival clouds. Google withdrew its complaint in November, but emphasized it was not retreating. Its complaint helped push the EC to launch a broader sector-wide investigation into cloud licensing and interoperability. Among other things, the EC is considering whether Azure and AWS should face new obligations for data portability and restrictions on bundling.
The U.S. Federal Trade Commission is also scrutinizing whether data egress fees, long-term contracts, and bundled licensing agreements hinder customers from switching providers, following industry complaints about opaque billing and technical barriers.
Meanwhile, in early 2025, the UK’s Competition and Markets Authority concluded that competition among Cloud Service Providers (CSPs) is insufficient, citing egress fees and restrictive licensing by Microsoft and Amazon as contributing to customer lock-in, with both companies controlling a significant share of the U.K. market.
Specifically, it found that Microsoft’s restrictive licensing and pricing practices give it the ability and incentive to partially foreclose competitors like AWS and Google, harming competition and consumers by raising effective switching costs and introducing friction throughout the cloud ecosystem. Key obstacles identified include high egress fees, significant market concentration, and technical limitations that make migrating workloads between providers challenging. With Microsoft and Amazon controlling up to 70–80% of the UK’s infrastructure‑as‑a‑service market, customers face limited alternatives and heightened operational risks when moving workloads away from these dominant platforms, amplifying the impact of licensing restrictions.
Microsoft has strongly contested these findings, arguing that regulators are unfairly targeting the company and that AWS and Google’s push for licensing reforms is “extraordinary and unprecedented.” The company insists its pricing strategies represent lawful competition and that its rivals remain capable of competing effectively. However, Microsoft’s rebuttals focus on broader competitive arguments rather than revealing the underlying contract language, leaving the specific terms driving these structural barriers opaque to agencies and the public.
Read our previous post: Are Clouds Too Sticky? Global Antirust Authorities Probe Lock-In Pricing Complaints.
What types of CSP customers can be harmed in California?
The harms are not theoretical. California’s economy runs on cloud infrastructure, and its mid‑market brands, platforms, and clinics are data‑intensive. If switching clouds triggers surcharges for the software an organization already owns, or if moving data incurs high egress fees, competition loses and costs climb.
The types of California businesses that could be harmed include:
- Sustainable consumer brands, especially mid‑sized operations engaged in global e‑commerce, rely on cloud portability to manage costs and foster innovation. These companies are highly sensitive to exit barriers and licensing surcharges, which can stifle their ability to adapt and compete.
- Online marketplaces and ecommerce platforms depend on cloud-based analytics, authentication, and logistics. Their complex architectures are particularly vulnerable to high data egress fees and proprietary APIs, which can impede efficient operations and increase costs.
- SaaS providers serving a broad range of clients may not be directly impacted by lock-in, but their customers are indirectly affected by hyperscaler terms that influence the cost and portability of the underlying platforms these services are built upon.
- Mission-critical service platforms, such as those handling large volumes of transactions or sensitive documents, are acutely exposed to portability and exit costs. These factors can have significant ripple effects on operational efficiency and customer experience.
- Fintech firms face unique challenges due to regulated data, multi-cloud risk management requirements, and the necessity for portability. For these companies, the ability to move between cloud providers is essential for controlling costs and managing risk.
- Genomics and precision health companies process massive data flows and advanced analytics. For them, the concept of “data gravity” is a tangible operational concern, where cloud exit costs and data transfer restrictions can become direct financial burdens.
Which of California’s cloud rivals and alternatives are at risk?
Lock‑in doesn’t only burden customers; it blunts business opportunities for California‑based competitors and infrastructure providers:
- Independent cloud storage providers that offer open, lower-cost storage solutions face challenges when exit costs and licensing surcharges make it difficult for customers to switch providers. These barriers can hinder growth and reduce competitive opportunities for such companies.
- Edge networking and Zero Trust platforms specializing in edge computing and security thrive on customer flexibility to mix and match cloud and network services. Restrictive licensing and high exit fees threaten this flexibility, undermining innovation and customer choice in the sector.
- Hybrid cloud infrastructure operators that provide backbone services for hybrid cloud strategies are impacted when licensing surcharges discourage moving workloads, such as Windows or SQL, across different providers. These costs limit the effectiveness and adoption of hybrid solutions.
- Multicloud integrators and service partners that help clients deploy and manage workloads across multiple clouds are harmed by discriminatory licensing that inflates costs on non-dominant platforms. This reduces the ability to offer diverse cloud options and restricts customer choice.
- Enterprise cloud file and data management providers are especially vulnerable to high egress fees and proprietary service hooks that slow and complicate migration. These obstacles can make exits costly and impede operational efficiency.
- Alternative infrastructure and decentralized computing models—such as companies building on-premises cloud computers or decentralized GPU marketplaces—rely on genuine portability. They are disadvantaged when dominant licensing rules and exit costs create barriers to moving workloads, limiting the adoption of innovative solutions.
California’s brand of innovation rewards agility. Contractual design should, too. Where a mid‑market retailer or a health‑tech lab cannot afford to re‑architect because exit costs are prohibitive, innovation stalls. Where a managed service provider, or MSP, can’t offer a viable multicloud path because licensing rules punish cross‑provider deployments, choice evaporates.
What can businesses do now?
Organizations currently facing this situation will want to document any harms caused by a cloud lock-in, such as delayed migrations, unexpected surcharges, and disabling of features on rival providers. These narratives matter both in government investigations and if a company wishes to explore private litigation.
But organizations who wish to avoid lock‑ins, cloud experts say, should “architect for exit,” not because you expect to leave tomorrow, but because portability gives organizations leverage today. Bolt these escape paths into contracts, they advise, e.g., negotiate egress‑fee caps, migration rights, and non‑discriminatory licensing terms for workloads that may run across multiple clouds. Treat data‑portability testing like any other resilience exercise; run it annually, validate your runbooks, and quantify potential egress liabilities as part of your standard risk posture.
These experts emphasize that an exit strategy must be embedded into cloud architecture from the beginning, with organizations proactively negotiating egress terms and designing for portability to prevent costly vendor lock‑in. For more on this, read The Importance of a Cloud Exit Strategy by Christian Siegers, writing for Architecture & Governance Magazine—and How Data Egress Fees Trap Enterprises in the Cloud by Osman Bin Abubakar, published at AZ Innovate Hub, which examines egress fees, lock‑in economics, and the need for proactive exit planning.
System architect Bence Hézső, CISSP, PMP, writing for the ISC2 association, likewise believes every organization needs a cloud exit strategy for reasons including data sovereignty, portability, security, and the freedom to adopt new technologies. He argues that reducing lock‑in requires a strategic commitment to open standards, along with multicloud and hybrid architectures that preserve long‑term flexibility. This approach ensures organizations not only meet their business and security objectives, but also maintains their “operational independence in the constantly evolving cloud landscape – all while remaining competitive, secure, and agile in today’s digital age.” Read Hézső’s article, Cloud Exit Strategies: Why and How to Avoid Vendor Lock-In.
Bakul Banthia, co‑founder of Tessell, agrees that vendor lock‑in can limit access to emerging technologies. Based in San Francisco, Tessell provides a fully managed, high‑performance Database‑as‑a‑Service (DaaS) platform designed to simplify cloud database management. Banthia writes that, as the technology stack shifts, cloud providers may not keep pace—and some may become obsolete if they fail to invest or evolve. For enterprises, cloud computing is indispensable, but choosing a provider capable of meeting future needs remains difficult. Banthia notes that organizations must continuously reassess whether their cloud provider’s capabilities align with their digital‑transformation goals, and that open approaches and multi‑platform strategies help reduce exposure to any single vendor’s limitations.
Simply put: switching providers is expensive. CSPs design their platforms as “walled gardens,” and once inside, it can be difficult to unwind tightly integrated services, Banthia writes. Egress fees are now standard and migration requires more than moving data—it involves re‑architecting systems, re‑mapping dependencies, and rebuilding operational patterns. For precisely these reasons, planning for an exit from the outset is not a sign of disloyalty; it is responsible architecture, Banthia advises. Read his article, Understanding the Risks of Cloud Vendor Lock-In.
While it is the subject of scrutiny in Europe, Amazon AWS appears to agree that switching cloud service providers should not be difficult. In its white paper, “Unpicking Vendor Lock-in: A Guide to Understanding and Mitigating Switching Costs When Changing Your Cloud Services Provider,” the company asserts that it “never wants to trap customers with lock-in tactics such as fixed-price, mandatory long-term contracts, or technical hurdles to changing CSP.” Instead, AWS says it aims to earn customer loyalty through innovation and service quality. “AWS wants customers to stay by choice, because AWS offers the broadest choice of the best cloud services,” the paper reads. It also notes several AWS tools and services that make data migration smoother and cost-efficient.
Conclusion
California’s tech industry helped invent the cloud, and now its future depends on ensuring competition and consumer choice. As competition authorities, courts, and tech leaders grapple with cloud lock-in and its consequences, the stakes have never been higher. The coming years will bring new legal standards, contract reforms, and potentially a more dynamic digital marketplace. For the innovators and businesses driving California’s economy — or economies across the country — attention to these developments is essential.
