If you’ve felt like every platform update this year has been circling the same idea, you’re not imagining it. Google, Meta, TikTok, and Apple have all been moving in the same direction: toward AI systems that don’t just assist your marketing, but increasingly run pieces of it, and toward discovery happening inside a conversation with an AI rather than a search bar or an app store category page. Taken one at a time, each update looks like its own isolated announcement. Lined up together, they tell a much bigger story about where mobile marketing is actually headed.
Below, we lay out the throughlines that have defined the year so far, then turn to July’s biggest mobile developments and what they signal for the months ahead.
The Year’s Throughlines
Discovery Is Moving Into the AI Answer
The single biggest shift of the year is where people go to find things. When OpenAI began testing ads inside ChatGPT back in February, it introduced a genuinely new kind of ad inventory: instead of bidding on a keyword, a rough proxy for what someone wants, brands bid on the context of a full conversation, a much richer signal of intent. Around the same time, Generative Engine Optimization (GEO) emerged as its own discipline, distinct from traditional SEO. AI citations, meaning mentions or recommendations from ChatGPT, Gemini, and similar tools, are now tracked as a standalone metric, and these engines favor structured, evidence-based content over keyword-stuffed pages.
For mobile specifically, this reshapes App Store Optimization. Users increasingly describe a need to an AI rather than typing a keyword or browsing a category, which creates a new optimization surface: visibility inside an AI’s answer. The job becomes influencing how a model interprets an app’s purpose and trustworthiness, not just its keyword ranking, even as the store itself remains where most installs still get decided.
The Platforms Want to Run the Operations
Across a spring run of major industry events, Google, TikTok, and Meta all delivered the same underlying message: set the goal, and the platform’s AI will handle the keywords, creative, and targeting underneath it.
At Google Marketing Live, Google confirmed that legacy Dynamic Search Ads will auto-upgrade to its AI Max system by September, a system that runs entire search campaigns automatically instead of requiring a marketer to set individual keywords, and introduced Ask Advisor, a conversational tool that proposes campaign structures from live account data. TikTok, per Moburst’s own coverage, pushed its Symphony AI suite from supporting creators to replacing parts of production, generating dozens of language variants of a script from a single take, and formalized its role as a search engine with brand-owned Search Hubs. Meta, meanwhile, was forecast to pass Google in global ad revenue for the first time in the history of digital advertising, driven by growing adoption of Advantage+ (its automated system for building and optimizing campaigns with minimal manual setup) and AI-generated creative pipelines.
The takeaway for practitioners: the skills that won paid media five years ago, granular keyword research and manual bid management, are being automated out of the job. The edge that’s left is upstream, in clean conversion data, well-structured feeds, and creative an AI system can actually learn from.
Apple Rebuilt the App Store Around AI
Apple’s Worldwide Developers Conference restructured how apps get discovered, sold, and classified, with the changes shipping alongside iOS 27 this fall. The pieces that matter most for mobile growth teams:
- A smarter Siri, but only through App Intents. The rebuilt Siri chains actions across apps using a semantic index of personal data, but it can only reach a third-party app through App Intents, a framework developers use to define which actions Siri is allowed to trigger. Apple issued a deprecation notice for the older SiriKit system, so apps that haven’t migrated won’t appear in cross-app Siri interactions.
- Creative Assets. Developers can now place custom images and video directly in product page headers and search results, submitted independently from app updates, turning store presentation into a live testing surface.
- Personalized Collections. The store now surfaces recommendations based on a user’s behavior instead of the same editorial picks for everyone, so post-install engagement and retention now feed back into how new users discover an app.
- The biggest subscription overhaul since 2016. Cross-developer bundles, group purchases, Volume Purchasing for enterprise and education, and Retention Messaging at the moment of cancellation together reshape how subscription apps are sold.
Because these land on a fixed fall deadline rather than a soft rollout, teams have a defined window to prepare rather than an open-ended one.
Measurement Tightened Around Consent and Trust
Quietly, the plumbing underneath all of this kept tightening. Google forced advertisers onto its newer Data Manager API, breaking any technical connection left inactive for 180 days and quietly staling remarketing audiences for anyone who hadn’t migrated. Meta, as its April updates showed, redefined what counts as a click, separating genuine link clicks from a renamed engagement category. The direction of travel is consistent: attribution is moving from tracking individual users toward measuring groups of similar users, using only data people have actually agreed to share, and that shift is now enforcement-backed rather than best-practice-only.
What July Signals for the Rest of 2026
If the first half of the year established the direction, July’s developments show how fast it’s now moving. These are the updates most likely to shape the remainder of 2026.
ChatGPT ads became a real channel, not an experiment
OpenAI expanded its ad pilot into the UK, Mexico, Brazil, Japan, and South Korea, and began testing multi-advertiser units that group several relevant ads into a single placement. Ad-tech partner Criteo reported that more than 2,000 brands are now running on ChatGPT through its platform. What began as a curiosity in February is now inventory with real scale.
What it spells: Expect conversational ad placements to move from test budgets to line items over the back half of the year, and expect early movers to lock in performance benchmarks before CPMs climb.
Google wired ads into the AI answer itself
The Gemini-powered formats from Google Marketing Live, Conversational Discovery ads, Highlighted Answers, and a Business Agent for Leads, moved from announcement into US testing. Highlighted Answers can place an ad inside an AI Mode recommendation list, and the Business Agent reads a site in real time to answer buyer questions inside the ad unit before handing off a pre-filled lead form.
What it means for brands: Your ad copy is no longer the whole message. The structured content on your site is ad copy now too, because the agent is reading from it live, which makes clean product data and clear site messaging a paid-media requirement, not just SEO hygiene.
Zero-click search stopped being a forecast
Roughly 68% of US Google searches now end without a click to any website, with AI Overviews correlating with a steep drop in click-through for top-ranking pages and referral traffic to publishers falling sharply. There’s a counterweight, though: a Similarweb study reported that when users encounter ChatGPT website recommendations, they’re about 2.5x more likely to visit the recommended page within seven days.
What it means: “Sessions from organic” is describing a shrinking world. The metric that matters for the rest of the year is citation share, being the source an AI quotes inside the answer, and the brands building structured, citeable content now will have a head start when AI citation rate becomes a standard dashboard line.
Meta lowered the barrier to entry
Meta shipped a one-click AI setup for the Pixel and Conversions API, removing the developer, cost, and maintenance overhead that used to make server-side tracking painful for smaller advertisers. Its AI business assistant in Ads Manager rolled out worldwide, and Threads opened static carousels and video ads globally.
What it spells: Threads inventory is new, comparatively uncluttered, and worth a test budget for app marketers whose audience has migrated there, treating the next few months as a cheap experimentation window before the inventory crowds and CPMs rise.
The regulatory ground shifted under the ad stack
Google’s updated Ads Terms of Service took effect July 1, reaffirming that advertisers remain liable for the assets its automation produces, the machine writes the headline, but you own it. In parallel, a remedies ruling in Google’s ad-tech antitrust case is expected imminently, with the DOJ pushing for divestiture, while TikTok’s ownership question moved toward resolution through the US joint venture established to satisfy the divestiture order.
What this means for brands: As more of the buying motion moves into automated black boxes, accountability stays with the advertiser and agency, and a potential reshaping of Google’s ad-tech stack rewards teams that diversify channels before they’re forced to.
Reddit kept pulling in AI-driven shoppers
With Reddit now a frequently cited source in AI answers and a place people go to sanity-check that advice against real human experience, the platform rolled out new native ad formats that fold user-generated content in alongside advertiser content. For app marketers, Reddit is becoming a meaningful mid-funnel surface where prospects validate a recommendation before they act, which makes an authentic presence there (not just paid placement) worth building into the plan for the rest of the year.
The Bottom Line
Line up the year so far, and the throughline is unmistakable: discovery is moving into AI-mediated surfaces, execution is moving into AI agents, and measurement is tightening around consent and traceability rather than raw reach. July didn’t change that direction, it accelerated it, turning what were announcements in the spring into live inventory, active tests, and hard deadlines. None of this displaces the fundamentals. The App Store still decides most installs today, retention still beats acquisition on unit economics, and these changes are shipping on fixed timelines rather than vague roadmaps. The teams that treat the iOS 27 rollout, the AI Max cutover, and the shift to AI-answer discovery as concrete deadlines, not background trends, are the ones who’ll be ahead of it.
Stay tuned for our next roundup. For more marketing insights, visit the Moburst blog or explore our full range of digital marketing services.
Related Articles
Digital Transformation Roadmap for Traditional Businesses
Moburst’s Mobile Marketing Digest: WWDC26 Edition
