Signal Scanner · REGULATION, STANDARDS & POLICY CHANGE · 5 October 2026

Competition Regulators Now Treat AI Assistants as Search Rivals Entitled to Google's Data and Android Access

EU and UK competition orders from July to September 2026 give AI assistants a route to Google's search data, Android access and choice-screen slots, so competition law, not AI statutes, sets AI distribution rules, with 2027 deadlines for assistant providers and device makers.

Produced using Shaping Tomorrow’s six-stage process for converting external change into decision-ready intelligence.
Scanned and filtered by AI. Stress-tested and judged by humans.

The consensus is that AI regulation means the AI Act. A quieter channel is competition law. In July the European Commission specified that “AI chatbots offering search functionalities are eligible to receive shared data” from Google (European Commission, 16/07/2026). On 23 September the UK Competition and Markets Authority proposed that AI assistants meeting technical and security criteria join Google's search choice screens (CMA, 23/09/2026). Neither measure assesses AI risk. Both decide which assistants count as search rivals and on what terms they reach users. Those eligibility tests, and a legal challenge Google filed on 29 September (Reuters, 29/09/2026), will set AI assistants' market access through 2028.

Signal Identification

A regulatory pivot. The orders rest on existing platform-competition powers, not AI statutes, and give AI assistants rights by classing them as search services. On the available evidence, the eligibility definitions are now the live regulatory object, and the CMA's are still being drafted.

Time horizon: 1-3 years (CMA decision by end-2026; EU search data from January 2027; Android access by August 2027; deferred hotword feature by August 2028)
access clocks run202620272028
Plausibility band: Medium-High
LowMediumHigh
Geographic / Jurisdictional Scope: Primary: the EU/EEA (Digital Markets Act orders on Google) and the United Kingdom (CMA digital markets regime). Spillover: other designated firms, the US and other regimes weighing AI-distribution remedies.
PrimaryEU/EEAUnited Kingdom
SpilloverOther designated firmsUSOther regimes
Sectors exposed:
AI assistant and chatbot providersSearch and search advertisingMobile device makers and app developersDigital publishersPrivacy and security functions

What's Changing

On 16 July the Commission specified how Google complies with the Digital Markets Act (DMA). An AI chatbot “that provides search-engine functionality in the EEA can qualify as a beneficiary, provided it clears a genuine-economic-activity test” (Bratby Law, 17/07/2026). Search data flows “from January 2027” and eleven Android features open “by 1 August 2027 at the latest” (Bratby Law, 17/07/2026). The Commission says the Android gap made rival assistants “less attractive to 60% of EU users who have an Android device” (European Commission, 16/07/2026).

The fines drew the headlines: €460 million for self-preferencing and €430 million over Play steering, with 60 days to comply (European Commission, 23/07/2026). Dialogue on applying the decision to “AI Overviews and AI Mode” continues. In the UK, the CMA would let AI assistants onto Google's choice screens, with views due by 9 October and a decision by year-end (CMA, 23/09/2026), after dropping the test that a service be marketed with general search as “a core and central part” of it (Bratby Law, 24/09/2026).

Two regulators set AI-assistant access on dated clocks

UK: CMA on Google search Filled: done. Hollow: still due. 3 Jun 2026 Publisher controls imposed 17 Jun 2026 Fair ranking, data portability 23 Sep 2026 Choice screen opens to AI 9 Oct 2026 Consultation closes End 2026 Final decision expected EU: Commission on Google Filled: done. Hollow: still due. 16 Jul 2026 Search data and Android orders 23 Jul 2026 Fines €460m and €430m Jan 2027 Search data flows to rivals 1 Aug 2027 11 Android features open 1 Aug 2028 Always-on hotword, deferred

Sources: CMA (23/09/2026); European Commission (16/07/2026 and 23/07/2026); Bratby Law (17/07/2026). Filled markers are steps taken; hollow markers are scheduled or pending.

Disruption Pathway

Stage one, to the end of 2026, is definition: the CMA settles who qualifies and the General Court receives Google's challenge. Stage two, 2027, is access: data flows from January on terms “fixed for a term of five years” (Bratby Law, 17/07/2026), and Android hooks open by 1 August 2027. Stage three, 2028 onward, is extension: if the Commission applies its self-preferencing decision to AI Overviews and AI Mode (European Commission, 23/07/2026), non-discrimination would reach Google's own answers. That step is this report's inference.

Stress concentrates in three places. Privacy: Google says the orders “force us to share people's private search history without sufficient anonymisation” (Reuters, 29/09/2026). Device security: the Cloud Security Alliance reads the Android order as “a mandated, standing interoperability surface that any qualifying third-party AI assistant can request equal access to” (Cloud Security Alliance, 17/07/2026). Eligibility edges: in the UK, “a significant proportion” of users must see an assistant as general search (Bratby Law, 24/09/2026). Two adaptations follow: assistant providers should log search-style usage now to evidence the tests, and budget for data and access costs instead of assuming free inputs.

Why This Matters Now

General counsel, regulatory affairs heads and strategy chiefs at assistant providers, publishers and device makers own this. Distribution plans usually assume defaults are bought or bundled; these orders add a regulated route: qualify as a search rival, then claim data, device hooks and screen slots. Eligibility, not model quality, becomes the gate, as Bruegel's Fiona Scott Morton says the policy intends: “The merits of an AI service, rather than whether it already owns a distribution channel, will determine its commercial success” (Bruegel, 01/07/2026). Boards should map each assistant against the EU and UK tests before the CMA decides this year.

Decision-action posture for this signal: Prepare — the eligibility tests are still being drafted and Google is in court, so mapping and cost modelling are justified now, with commitments held for the CMA decision and the court's handling of the challenge.

Counter-Argument

The strongest objection is that the access is paper access. Google is appealing, saying the orders “weaken vital security protections on Android” (Reuters, 29/09/2026). The CMA says assistants “could in theory be able to be listed” (Bratby Law, 24/09/2026). A long case, a narrowed order or a choice screen that no assistant qualifies for would leave only drafting.

Delay still leaves the vocabulary in place. Both regulators have written AI assistants into the definition of a search rival, and the AI Overviews dialogue puts Google's own AI search features inside the same non-discrimination test. This report's reading is that even a narrowed outcome makes eligibility, not model risk, a threshold AI assistants must prove.

Implications

The direction looks durable and the detail uncertain. On the available evidence, competition law is becoming the binding layer on how AI assistants reach users, a subject AI statutes do not address. Bruegel calls the Android specification “a substantive and well-targeted intervention” (Bruegel, 01/07/2026), and the CMA ties its step to people using assistants as their search service (CMA, 23/09/2026). The inflection window runs from the CMA decision in 2026 to the Android deadline in 2027, with 2028 for any extension to AI Overviews. Assistants that can evidence search-like use stand to gain; providers below the thresholds and incumbents holding exclusive device hooks stand to lose.

Early Indicators to Monitor

Disconfirming Signals

Strategic Questions

Keywords

AI assistants; Digital Markets Act; search choice screen; CMA strategic market status; search data sharing; Android interoperability; AI Overviews; self-preferencing; competition remedies; eligibility tests

Bibliography

Source tiers: Tier 1, governments, regulators and intergovernmental bodies. Tier 2, think-tanks, academic institutes, major consultancies and quality data providers. Tier 3, quality journalism and specialist trade press. Tier 4, vendor, company and practitioner sources, used only as directional corroboration.

How this was made

The Decision-Grade Intelligence Method

Shaping Tomorrow’s six-stage process for converting external change into decision-ready intelligence.

Scan · Filter · Stress-test · Judge · Trigger · Actionable Intelligence

See full methodology → · How we verify → · Track record →

This Signal Scan is one component of a Shaping Tomorrow intelligence cycle on Regulation, Standards & Policy Change. Every monitored topic also includes monthly Change Trackers that assess what is materially shifting in momentum, and a quarterly Decision Intelligence briefing that converts the cycle into board-level strategic implications. Cycles may also close with Action Triggers, an end-of-cycle synthesis that reads across the three formats and surfaces patterns no single deliverable carries on its own. Continuous intelligence on this topic is available as a Single-Topic or Multi-Topic programme.

Discuss a Tailored Programme

Prepared by Shaping Tomorrow: 5 October 2026