EU Commission Declares DMA Fit for Purpose in First Statutory Review and Pivots Enforcement Toward AI and Cloud
Brussels' first review of the Digital Markets Act concludes the law needs no overhaul, declines to designate generative AI as a new core platform service, and shifts focus toward cloud gatekeeper probes and a Google Android AI specification.
Overview
The European Commission published its first statutory review of the Digital Markets Act on April 28, 2026, concluding that the regulation “remains fit for purpose and has opened up new opportunities for businesses and developers,” according to the European Commission. The review, required by the DMA’s own text within two years of its core obligations becoming applicable, declines to add generative AI as a new core platform service and leaves the gatekeeper list unchanged, while signalling that the next phase of enforcement will concentrate on cloud computing and on shoring up AI competition through existing legal hooks.
As The Machine Herald previously reported, the Commission was widely expected to use the review to decide whether to widen the DMA’s reach into AI and cloud. The verdict is more cautious than the most ambitious calls from civil society and consumer groups, but it sits alongside two parallel actions, a cloud designation track and a fresh Android AI specification, that together signal the Commission intends to deploy the law it has rather than rewrite it.
What We Know
The Commission’s verdict on the DMA
The review found that the DMA “has thus far effectively contributed to the core objectives of making digital markets in the EU fairer and more contestable,” and concluded that it “remains fit for purpose and does not need substantial revision or modification,” according to TechPolicy.Press. The Commission’s evaluation page lists tangible outcomes for end users, including the ability to transfer data when switching between services and devices, and the choice of alternative search engines and browsers over manufacturer defaults, as the European Commission noted.
For business users, the Commission cited new openings for connected device manufacturers seeking interoperability access, the launch of third-party app stores following ecosystem openings, and new messaging apps enabled by interoperability obligations, per the European Commission. Euronews highlighted that WhatsApp users can now message people on competing platforms as a concrete example of the interoperability obligation in practice.
The DMA has been applicable since May 2023, with the current designations in force covering seven companies: Alphabet, Amazon, Apple, Booking.com, ByteDance, Meta Platforms, and Microsoft, according to coverage of the review by The Register and the broader analysis in TechPolicy.Press.
What is not changing
Despite calls from more than 450 consultation respondents to broaden the law, the Commission opted to keep the core platform services list unchanged in this review, TechPolicy.Press reported. Three areas of restraint stand out:
- The Commission will not add generative AI as a new core platform service at this time. Instead, it plans to assess whether some AI services should be designated under the existing virtual assistants category, which currently has no designated services, according to TechPolicy.Press.
- The Article 7 interoperability obligations will not be extended to social networking services, TechPolicy.Press noted, with the Commission stating it is “too soon to draw conclusions” on the impact of messaging interoperability so far.
- The list of designated gatekeepers and the thresholds used to designate them remain unchanged, per TechPolicy.Press.
Where enforcement is moving next
The Commission’s restraint on the legislative side is paired with a clear shift in enforcement focus. Teresa Ribera, the Commission’s Executive Vice-President for a Clean, Just and Competitive Transition, signalled the direction in remarks reported by The Register and in a statement carried by Reuters wire coverage that “The DMA was designed to be future-proof and adapt to emerging challenges, for example in AI and cloud.”
On cloud, three market investigations launched in November 2025 are assessing whether Microsoft Azure and Amazon Web Services should be designated as gatekeepers, with designation decisions expected by November 2026 and an investigation conclusion scheduled for May 2027, according to TechPolicy.Press. The Commission is examining whether Article 6(7) of the DMA could be applied to cloud services to address “obstacles to interoperability between cloud computing services, limited or conditioned access for business users to data, tying and bundling services,” the TechPolicy.Press analysis quoted from the review.
On AI, rather than reaching for a new core platform service category, the Commission opened a specification proceeding under Article 6(7) targeting Google. The European Commission framed the action as part of the DMA’s normal enforcement cycle. The Register reported that the proceeding, initiated in January 2026, sent preliminary findings to Google on April 28, 2026, with a public consultation deadline of May 13, 2026.
The Commission’s preliminary view, as summarised by The Register, is that Google should let competing AI services activate via custom wake words, achieve integration parity with Gemini on Android, and perform user-facing tasks such as sending emails, ordering food, and sharing photos with comparable ease. Ribera said in The Register: “Today’s proposed measures will give more choice to Android users about the AI services they use and integrate in their phone, including from the vast range of AI services that compete with Google’s own AI.”
Google pushed back. Clare Kelly, the company’s senior competition counsel, told The Register that the proposal amounts to “unwarranted intervention” that “would strip away that autonomy, mandate access to sensitive hardware and device permissions.”
Industry and civil society reactions
The overall reception has split along familiar lines. Apple criticised the report for not adequately accounting for impacts on privacy, security, and innovation, citing risks tied to alternative app distribution channels, according to Reuters wire coverage referenced in the TechPolicy.Press analysis. On the other side, some stakeholders called for “stronger use of sanctions, including higher fines” and “interim measures to more quickly tackle anticompetitive behavior,” TechPolicy.Press reported.
What We Don’t Know
Several important questions are left open by the review.
First, the cloud designation track is well underway, but the timetable TechPolicy.Press reports points to designations “expected by November 2026” and a final investigation conclusion in May 2027. Whether Microsoft Azure or Amazon Web Services will ultimately be subjected to the full gatekeeper regime, and which obligations will apply, will not be settled in this review cycle.
Second, while the Commission opened a specification proceeding for Google on Android AI access, the Article 6(7) framework relies on case-by-case implementing decisions rather than blanket rules. The May 13 consultation deadline reported by The Register is the next visible milestone, but the timing and content of any final implementing decision are not specified.
Third, the review does not address ongoing transatlantic friction over EU tech enforcement, an absence that critics highlighted. Aline Blankertz of Rebalance Now noted to TechPolicy.Press that the review remained “silent on pressures from the US to weaken enforcement.”
Analysis
The choice not to broaden the DMA in its first review is consequential. By keeping the core platform services list intact, the Commission avoids a legislative renegotiation that could have taken eighteen months or more, and it preserves the law’s existing case backlog and procedural cadence. By using Article 6(7) specification proceedings for AI competition concerns on Android and a separate market investigation track for cloud, the Commission is opting to stretch the existing framework rather than amend it.
That approach has trade-offs. Specification proceedings are narrowly tailored to designated gatekeepers and identified bottlenecks, which makes them flexible but slow to scale across the AI sector. The virtual assistants category, which TechPolicy.Press notes has no current designations, gives the Commission a future avenue for AI-specific designations without rewriting the law, but only if a service meets the existing thresholds and definitions.
For the seven currently designated gatekeepers, the practical message of the review is continuity. The legal regime that defines their obligations will not change in the near term. What is changing is where the Commission will spend its enforcement attention. Cloud and Android AI access are now the visible priorities, alongside the longer-running cases on app store anti-steering and pay-or-consent advertising models that drove the headline fines of 2025. For potential new entrants, including European AI providers, the Article 6(7) specification on Android is the first concrete test of whether the DMA can deliver competitive access to a hardware ecosystem in a market where the rules of the road are still being written.