Will the EU’s DMA Open Android to Third-Party AI?

Will the EU’s DMA Open Android to Third-Party AI?

Every time a phone perks up to “Hey Google,” a quiet, high‑stakes race begins over who controls the next action, the next data point, and the next habit that shapes a user’s digital life. That single phrase sets the path for search, directions, messages, and even shopping, channeling intent through one assistant’s lens.

Now the European Commission has stepped in, arguing that Android’s gatekeeper status comes with duties: make the door wider, or at least stop bolting it from the inside. The question is not only whether rivals can speak, but whether the device will listen with equal speed, context, and reliability.

Nut Graph

Under the Digital Markets Act, Google’s mobile platform must avoid steering users toward its own AI by design. Regulators say the default invocations, exclusive context pipes, and privileged hardware hooks tilt the field for Gemini. In a public consultation, the Commission framed parity across three fronts: invocation, context, and resources.

The stakes extend beyond competitive fairness. Assistants now sit at the operating system layer, where invocation latency and cross‑app execution decide utility. As one draft remedy puts it, the goal is “standardized, open interfaces for assistants” so that choice does not cost capability.

Body

Consider two chokepoints that drew scrutiny. The long‑press gesture on the navigation bar routes to Circle to Search and Gemini, enabling on‑screen overlays and instant actions. Meanwhile, always‑on wake word detection defaults to “Hey Google,” with no supported pathway for third‑party registration. Add AppSearch access limited to the default assistant, and an alternative app begins the race several steps behind.

Google counters that unfettered access could raise security risks and integration costs. “Opening sensitive permissions and low‑level hardware paths without tight controls threatens privacy and reliability,” the company argued in responses, warning of fragmentation if OEMs juggle inconsistent implementations.

Regulators point to precedent. When the EU challenged Apple Pay’s NFC exclusivity, access expanded industry‑wide, though national variations complicated rollout. The same logic now guides Android: neutral device controls, fair access to context, and predictable resource allocations. Developers echo the practical needs: stable APIs, clear quotas, audit trails, and consent flows that users actually complete. OEMs weigh compliance risk against new differentiation—health, finance, and enterprise copilots with on‑device data—if invocation and context are opened.

Conclusion

The path forward favored concrete milestones: user‑selectable wake words, remappable gestures, multi‑assistant registration, and permissioned on‑screen capture with redaction. Success hinged on documented access to audio pipelines, sensors, and ML accelerators, plus telemetry caps and incident reporting to keep trust intact. If Google published neutral interfaces, if OEMs shipped clear settings and wake word managers, and if third‑party AIs optimized for latency and battery, Europe would have measured progress through time‑to‑respond, consent completion, and real user switching. The door to choice had not swung by rhetoric; it moved when invocation, context, and resources aligned under enforceable rules.

Subscribe to our weekly news digest.

Join now and become a part of our fast-growing community.

Invalid Email Address
Thanks for Subscribing!
We'll be sending you our best soon!
Something went wrong, please try again later