How the UK Retreated on Cloud and Called Its Local Media Band-Aid a Plan

Courtney C. Radsch, PhD, is a journalist, scholar and human rights advocate who currently directs the Center for Journalism & Liberty at Open Markets Institute and is a non-resident fellow at the Brookings Institution. She serves on the board of Tech Policy Press.

Last week, the UK government made two announcements about the information economy that few saw as connected, but which portend little change in the underlying power that US tech platforms play in Britain. On March 26, Culture Secretary Lisa Nandy unveiled Amplify, the government's local media action plan, the first in a generation. A few days later, on March 31, the Competition and Markets Authority (CMA) announced the outcome of its three-year investigation into the cloud computing market. Both announcements came packaged in the language of decisive action, but neither delivered what it promised.

Cloud computing

The gap between what the CMA found and what it did is deeply disappointing given hopes for its potential to meaningfully address strategic market status and the very obvious dominance of American hyperscalers in the cloud market. This was particularly surprising given the growing awareness about the threats that this dependence creates for digital sovereignty amid increased American pressure and belligerence.

When Ofcom referred the UK cloud market for investigation in October 2023, the reason for concern was straightforward: Amazon and Microsoft together control between 70 and 80 percent of UK cloud infrastructure (with Google a distant third). The way they structure licensing, egress fees, and interoperability makes it expensive and difficult for customers to go anywhere else. The investigation confirmed all of this (which we have also highlighted extensively, including the implications for investigative journalism), with the CMA's July 2025 final report concluding that competition in the market was not working well and that both companies held significant unilateral market power. It recommended the board consider designating both Amazon and Microsoft with Strategic Market Status, a designation under the Digital Markets, Competition and Consumers Act that would give the regulator binding powers to impose conduct requirements and structural remedies.

The board blinked. What emerged last week was not what the investigation recommended. Amazon got no designation, no formal probe, no binding requirements—just voluntary commitments on egress fees and interoperability, negotiated in private and announced with press releases. Microsoft got a new SMS investigation, but not into cloud, where the competition concerns were documented and severe. It got a probe into its business software ecosystem (Word, Excel, Teams, Copilot) that won't even reach a provisional finding for another nine months. The cloud inquiry chair had already resigned a month earlier over the “glacial” pace of reform and concerns about the agency's independence. Mark Boost, CEO of UK cloud provider Civo, summed up concerns about the outcome, calling it "a repeat of the provisional decision, but with softer edges." The former deputy government Chief Information Officer called it "intent, but not urgency."

The lack of urgency matters not just in cloud markets but for the economy writ large. Last year, the Open Markets Institute and the Center for Journalism and Liberty submitted formal comments to the CMA's SMS investigation into Google's search services, arguing that behavioral remedies alone are insufficient to address structural dominance. The history of Google's conduct in France, where the Autorité de la Concurrence spent years chasing the company for failing to comply with its own commitments, illustrates why. The CMA's cloud outcome is a live demonstration of why voluntary commitments from companies with 30-40 percent market share each, reviewed in six months at the regulator's discretion, are unlikely to be meaningful remedies. They are deferrals dressed up as progress and pose broader threats to the information ecosystem that the CMA failed to grapple with.

As my colleague Karina Montoya and I documented in a 2024 paper on cloud concentration and investigative journalism, the hyperscaler dominance the CMA identified is a press freedom problem as much as a competition problem. News organizations, including investigative outlets, have become structurally dependent on infrastructure controlled by a small number of companies whose commercial interests are not aligned with the public interest journalism those organizations produce and whose excessive fees, high switching costs, and lock in makes it challenging to find alternatives. (We are working on alternatives through the Journalism Cloud Alliance.)

The independence question has a name. The CMA's permanent chairman is Doug Gurr, who spent sixteen years at Amazon, most recently as its UK country manager. He recused himself from the cloud case, but the agency he chairs runs on Amazon's servers, and the enforcement posture of the watchdog under his leadership is being watched closely. As Tech Policy Press reported, concerns extend beyond his Amazon background to his broader approach to enforcement.

Local media

Meanwhile, across Whitehall, the Department for Digital, Culture, Media & Sport (DCMS) announced that it would save local journalism with a fund and by helping it “adapt” to changing audience consumption habits. It did nothing to address the role that cloud dominance plays in undermining local journalism. (Apparently these departments do not talk to each other.)

Amplify: The Local Media Action Plan is not without value. Tripling community radio funding to £1 million a year is great (though the failure to designate a sufficiently independent oversight body to disburse funds is not great). The commitment to future-proof the BBC's Local Democracy Reporting Service is important and worth doing. including a Local News Fund of up to £12 mllion, which could be an oasis of support (assuming a more robust vision of its independence from government is adopted).

But the government's own data makes the gap between diagnosis and remedy stark. At least 293 local newspapers have closed since 2005, with 4.4 million people now living in news deserts, according to Public Interest News Foundation. Local news publisher advertising revenue collapsed from £1.68 billion to £0.33 billion between 2011 and 2024.. The plan acknowledges that AI summaries drain traffic without compensation, that platforms suppress publisher content behind opaque terms violations, that 81% of AI news responses contain accuracy or sourcing problems. The proposal from the government’s plan to address the platform layer, which is where the damage is actually happening, is to ask the industry to explore voluntary options and monitor the situation.

We’ve seen where “voluntary” and “monitoring” get us. The tech corporations are bigger, wealthier, and more entrenched than ever. This mealy set of responses in the face of acknowledged crisis is a policy choice that will have repercussions for years to come. The same logic that let Amazon walk away from the cloud investigation with a handshake governs the approach to local journalism: trust the dominant players to behave themselves, frame accommodation as pragmatism, and call the whole thing a package of actions.

New teeth, no bite

Our letter to the CMA on Google's search services argued that we cannot address dominant digital companies as players in isolated markets because search powers adtech, which powers AI, which powers devices, which powers search again, in a self-reinforcing loop that behavioral nudges cannot break. Furthermore, the threats from platform concentration run deeper than advertising revenue, reaching into the infrastructure layer itself, where the same companies the CMA declined to formally designate in cloud are also the landlords of the digital systems that newsrooms depend on to function. The infrastructure layer and the content layer are not separate problems, but rather the same problem at different points in the stack.

Britain cultivated sharper new regulatory teeth for exactly this moment. The question is why it keeps smiling at the companies it was supposed to bite. And who pays the price while the meetings and consultations and voluntary (non)commitments continue? Voluntary commitments from the most powerful companies in the world are not a solution for market dominance, nor are they effective tech or media policy while millions live in communities with no one left to cover their council meetings, government procurement, or other important local issues.