Apple Intelligence Search Is on 200 Million Devices. Here's How Apple Picks What to Show.
Apple confirmed 2.5 billion active devices in its January 2026 earnings, and the slice running Apple Intelligence keeps growing every quarter. iPhone 16 outsold iPhone 15 in every market where Apple Intelligence was live, according to Tim Cook on the same earnings call. That is hundreds of millions of pockets where Siri now answers with summaries, ChatGPT replies, and on-device AI before anyone touches Google.
Most SEO teams are not optimising a single thing for it.
This post is what Apple has actually shipped, how the surfaces behave when you put them through real-world queries, and what I think you should do about it. To be clear up front: this is a strategic primer, not a controlled study. Apple does not publish a Search Console. There is no clean ranking dataset to crunch. What you'll find below is built from Apple's own documentation, public reporting, and informal testing on the surfaces themselves.
What "Apple Intelligence search" actually means
There is no single Apple search engine to rank on. There are five surfaces, and they behave differently.
Siri with ChatGPT integration. Since iOS 18.2, Siri can hand off "world knowledge" queries to ChatGPT after asking the user. Apple and OpenAI announced the partnership in June 2024 and shipped it through the autumn updates. The user sees a ChatGPT-styled card inside Siri, with citations sometimes, sometimes not.
Apple Intelligence summaries. When users get long emails, Safari articles, or notifications, Apple Intelligence summarises them on-device. This is not search, but it changes how content is consumed before it gets clicked.
Spotlight. The home-screen pull-down search now blends app content, contacts, web suggestions, and contextual results. App Intents (the developer framework Apple opened up to surface in-app actions) are starting to show up in Spotlight results.
Safari intelligent search. Safari's smart search bar uses Apple's own ranking signals plus Google's web results. Bloomberg's Mark Gurman has reported that Apple is building an internal system codenamed "World Knowledge Answers" to power AI search across Siri, Safari, and Spotlight, with Google's Gemini reportedly involved in the deal. That is a 2026 story to watch closely.
Apple Maps and Apple Business Connect. For local intent, Maps is the surface. Business listings managed through Apple Business Connect appear in Maps, Spotlight, Safari suggestions, and Siri local answers.
If you want a wider view of how this fits into the broader move past Google, my piece on search everywhere optimization beyond Google covers the full landscape.
The reach numbers, with the receipts
Apple does not publish a clean "Apple Intelligence devices" figure. Here is what we can verify.
Apple reported 2.5 billion active devices in early 2026, up from 2.35 billion the year before. Apple Intelligence requires an iPhone 15 Pro, iPhone 15 Pro Max, or any iPhone 16 model, plus iPads and Macs with M1 or later. The iPhone 16 line shipped over 100 million units in its first six months based on industry analyst estimates, and the iPhone 15 Pro and Pro Max sold at a similar pace through 2024. Add eligible iPads and Macs and you are comfortably in the 200 to 300 million range of Apple Intelligence-capable devices in active use as of Q1 2026.
The other number that matters: Google paid Apple roughly $20 billion in 2022 to be the default search engine in Safari, per court documents revealed during the DOJ antitrust case. Judge Amit Mehta ruled in August 2024 that the deal violated antitrust law. The remedies issued in September 2025 limited exclusivity but did not block the payments outright. So Apple still has a financial reason to keep Google in Safari, but the lock-in is loosening, and Apple is clearly building its own AI search alternative behind the scenes.
That is the strategic backdrop. Apple has the install base, the cash to build, and a regulatory push to reduce Google dependence. Whatever Apple Intelligence search looks like in 18 months, it will not look like a Google query box.
How the five surfaces actually behave
Here is what you can observe by putting different intent types through the surfaces on a current iPhone running iOS 18.x.
Informational queries lean hard on ChatGPT
Ask Siri something like "how long do car batteries last" or "what is creatine loading" and most of the time Siri will offer to hand off to ChatGPT, or will do it automatically depending on the user's settings. The ChatGPT response inside Apple Intelligence is almost identical to what you get on chatgpt.com with browsing turned on, but with a key difference: clickable source citations are inconsistent. Sometimes you get them, often you get plain prose with no obvious way to verify or click through.
This matters. If your content is being used as a retrieval signal but not cited, you get the cost of being scraped without the click. The fix is the same as for any LLM surface: clear factual statements, schema markup, citable structure. I covered the playbook in LLM optimization: how to get your brand into AI answers.
Navigational queries route to Spotlight or Safari, not ChatGPT
For brand and site queries, Siri rarely offers ChatGPT. It either opens the app, surfaces a Spotlight result, or feeds into Safari's smart search and lets Google handle it. This is the closest to "normal" SEO behaviour on the platform. If you rank well on Google for your brand, you are fine here. But if you have an iOS app, the app shows up faster than your website in Spotlight a clear majority of the time.
Worth flagging: App Store presence affects this. Brands without an iOS app fall back to web results every time. App Intents implementation, which Apple opened up to developers for richer Siri and Spotlight integration, is now a real ranking surface, not just a UX nice-to-have.
Local queries are dominated by Apple Maps, not Google
This is the most under-discussed shift. For "emergency plumber Melbourne", "best ramen Brisbane", and similar local queries on iPhone, Siri pulls directly from Apple Maps. The top three results are Maps listings. Google does not appear unless the user explicitly asks Siri to search Google.
If you have a local business and your Apple Maps listing is incomplete or unclaimed, you are not in the consideration set on iPhone. Full stop. Apple Business Connect is free, and the listing flows through to Maps, Spotlight, Safari suggestions, and Siri local answers. Apple's own marketing materials cite a 31% lift in clickthroughs from action links on Business Connect listings, though Apple does not publish the underlying methodology, so take the exact figure with a grain of salt.
If you want the deeper logic on how local ranking has shifted overall, my piece on local SEO 2026: popularity signals over prominence walks through the modern signal mix.
Transactional queries are messy
Mixed Spotlight cards, Apple Wallet promos, ChatGPT product summaries, and Google Shopping results in Safari. No clean pattern. ChatGPT through Apple Intelligence sometimes recommends specific brands, sometimes refuses. Apple's own Wallet and Pay integration occasionally jumps in.
For now, if you sell physical products to consumers, your job is the same as everywhere else: be everywhere. Google Shopping, your own product page schema, ChatGPT visibility, App Store presence if you have an app. There is no Apple-specific shortcut yet.
Apple's answers are shorter and more cautious than Google's
Across enough queries, you start to notice that Apple-side answers (Siri direct or ChatGPT handoff) are noticeably shorter than Google's AI Overview for the same query. Apple Intelligence refuses or punts to web search far more often. It says "I don't have that information" or "would you like me to search the web" with much higher frequency than Google AI Overviews do.
For SEOs this is good news. It means the Apple surface is hungry for clear, sourceable content and the bar to be cited is lower than on Google's AI Overviews, where the competition is brutal. If your content is well structured and authoritative, Apple's surface is one of the easier AI answer engines to land in right now.
How Apple's answers differ from Google's
A few patterns worth flagging if you compare the same query across both.
Apple favours Wikipedia and official sources at a higher rate. For research-style queries, Apple's ChatGPT handoff tends to cite peer-reviewed databases and Wikipedia. Google's overview pulls from a broader mix of commercial blogs and content sites. That tracks with OpenAI's known retrieval preferences, but it is sharper inside Apple's surface.
Apple under-indexes on freshness. For news-adjacent queries, Apple's answers are noticeably staler. Google wins almost every "what happened today with X" query. If your content depends on freshness signals, Google still owns that intent.
Apple respects the user's location more aggressively for local intent. Even when the user does not say "near me", Siri assumes local intent if the query has any location-able verb (find, get, book, buy near). Google is less willing to assume.
Apple silently strips brand recommendations on sensitive queries. For "best [supplement]" or "best [medication for X]" queries, Apple Intelligence almost always declines to recommend specific brands and instead suggests the user consult a professional. This is a guardrail. If you are in a regulated space, Apple will not be a brand awareness channel for you.
Practical optimisation tactics for Apple Intelligence
Here is what I would actually do this quarter if Apple visibility matters to your business.
1. Claim and fully build out Apple Business Connect
If you have any local component to your business, this is the single highest-leverage move. The Apple Business Connect platform is free, listings flow into Maps, Siri, Spotlight, and Safari, and the bar is lower than Google Business Profile. Add photos, hours, action links, attributes, and category information.
2. Make your content easy for ChatGPT to retrieve and cite
Since the bulk of informational answers route through ChatGPT inside Apple Intelligence, you optimise for ChatGPT and you cover the Apple surface at the same time. Clear factual statements early in the page, definition-style paragraphs, FAQ schema, and natural citation hooks (specific numbers, named studies, dates). My Bing SEO and ChatGPT 2026 guide covers this in depth, and most of it transfers directly.
3. Implement App Intents if you have an iOS app
If you have an iOS app, work with your developers to expose key actions through Apple's App Intents framework. This is what lets Siri, Spotlight, and Apple Intelligence surface in-app actions and content. In navigational testing, apps with rich App Intents implementation jump above their website in Spotlight results consistently.
4. Get your schema clean
Apple's surface, like every AI surface, leans on structured data. Article schema, FAQPage schema, LocalBusiness schema, Product schema. None of this is exotic, but plenty of sites have it half-implemented or wrong. Run a check.
5. Build for multi-surface visibility, not just Google
Apple is one surface. ChatGPT is another. Gemini, Perplexity, Bing Copilot, TikTok search, YouTube. The brands winning right now are visible across all of them with consistent positioning. If you want a deeper read on getting cited by AI engines, LLM optimization: how to get your brand into AI answers is the most-read post we have on it.
6. Track what you can
Apple does not give you a Search Console. You will not get clean keyword data. What you can do: monitor referral traffic from Spotlight (it shows up as a direct visit usually), watch your Apple Maps listing analytics inside Business Connect, and run periodic manual queries on a real iPhone for your priority terms. Set a calendar reminder. Quarterly is fine.
7. Watch for "World Knowledge Answers"
If Bloomberg's reporting is accurate, Apple's own AI search system rolls out through 2026 across Siri, Safari, and Spotlight. When that ships, the rules change again. Stay close to Apple's developer documentation and Search Engine Land's Apple platform coverage for early signals.
The honest summary
Apple Intelligence search is not a single ranking system. It is five surfaces stitched together with ChatGPT in the middle, Apple Maps doing the heavy lifting on local, and Apple's own AI search engine being built in the background. The reach is already big. It is going to get bigger.
Most SEO teams are not paying attention. That is the opportunity. The optimisation tactics are not exotic: claim Apple Business Connect, get your schema right, write for AI retrieval, build App Intents if you have an app, and stay multi-surface. None of that is new advice. It is just that almost nobody is doing it for Apple specifically.
The patterns will shift as Apple ships its own AI search. The principle will not: if you are easy to retrieve, easy to cite, and present on the surfaces your buyers actually use, you win regardless of whose search engine they are using.
