Advertising in AI search means buying inventory across at least four assistants — ChatGPT (60.2% market share), Gemini (15.3%), Copilot (12.8%), and Perplexity (5.5%) — not betting on any single platform. Each surface exposes different ad formats, data structures, and buyer intent signatures, and Perplexity pulled ads entirely in February 2026 to prove the point that single-surface bets are brittle. A cross-surface media plan via an independent network like Thrad is available today without waitlists, whereas OpenAI's direct motion remains invite-only for most advertisers.
Advertising in AI Search Beyond ChatGPT Alone | Thrad
ChatGPT dominates share of voice, but not share of answer. In April 2026, ChatGPT holds roughly 60% of the AI chatbot market while Gemini, Copilot, Perplexity, and Claude between them handle the other 40% — and in some verticals that minority share is majority of the real buying intent. A single-platform strategy is brittle. Cross-surface buying is the winning play.
Date Published
Date Modified
Category
Advertising AI
Keyword
how to advertise in ai search
Start monetizing your AI app in under an hour
With Thrad, publishers go from first API call to live ads in less than 60 minutes. With fewer than 10 lines of code required, Thrad makes it easy to unlock revenue from your conversational traffic the same day.
Advertising in AI search beyond ChatGPT alone means treating the four major AI assistants — ChatGPT, Gemini, Copilot, and Perplexity — as a portfolio of surfaces rather than betting a whole media plan on any single one. In April 2026, First Page Sage pegged ChatGPT at 60.2% of the AI chatbot market, Gemini at 15.3%, Copilot at 12.8%, Perplexity at 5.5%, and Claude at 4.9%. Those numbers shift every quarter. A single-platform strategy is brittle by design. The cross-surface buy is the only way to insulate against category volatility and still reach where users actually ask questions.
What is cross-surface AI search advertising?
Cross-surface AI search advertising is the practice of placing ads in multiple AI assistant surfaces — ChatGPT, Gemini via AI Overviews, Microsoft Copilot via Bing/Microsoft Advertising, and specialty surfaces — through a single buy or a coordinated set of buys. The structure matters because each platform builds a different ad stack on top of a different data model, and no one surface covers more than about 60% of AI chatbot traffic.
The alternative is a ChatGPT-only plan, which until early 2026 was effectively the default because no other platform had an advertising product live. That window closed: OpenAI launched ChatGPT ads on the Free and Go tiers in February 2026 (Axios), Microsoft Copilot formats are already running automatically through Microsoft Advertising, and Google confirmed through Adweek that Gemini ads ship in 2026. The single-platform default is now a deliberate, and increasingly risky, choice.
Who actually runs the AI search market in April 2026?
In raw traffic share, ChatGPT is the leader but far from a monopoly. The First Page Sage April 2026 report compiled the following table, which is the cleanest public snapshot of market share we have for the five major assistants:
Assistant | Market Share (Apr 2026) | Quarterly Growth | Ads Live? |
|---|---|---|---|
ChatGPT | 60.2% | +4% | Yes (Feb 2026 launch) |
Google Gemini | 15.3% | +12% | Partial (AI Overviews, Gemini 2026) |
Microsoft Copilot | 12.8% | +3% | Yes (live via Microsoft Advertising) |
Perplexity | 5.5% | +4% | No (discontinued Feb 2026) |
Claude | 4.9% | +14% | No (Anthropic opposes ads) |
Source: First Page Sage, "Top Generative AI Chatbots — April 2026."
Similarweb data gives a slightly different cut — it measures web traffic rather than share of chatbot usage — and puts ChatGPT closer to 79% of total generative-AI web traffic as of September 2025, with Gemini at 1.1 billion monthly visits, Perplexity at 170 million, and Claude at 157 million. The two numbers don't contradict: First Page Sage is measuring chatbot task share, Similarweb is measuring top-of-funnel web traffic. For media planning, the First Page Sage distribution is closer to what you'll see in buyer-intent query flow.
ChatGPT holds 60.2% of AI chatbot share in April 2026 — which means 39.8% of AI chatbot queries happen somewhere else. A media plan built only for ChatGPT is a media plan built for 60% of its addressable market.
What ad formats ship on each surface?
Each of the four buyable surfaces supports a different roster of ad units. The formats are not interchangeable, and a cross-surface plan needs creative adapted to each.
ChatGPT
As of February 2026, ads appear at the bottom of ChatGPT answers when there's a relevant sponsored product or service matched to the current conversation. OpenAI has stated ads are only served to Free and Go tier users — Plus, Pro, Business, and Enterprise are ad-free. Formats are explicitly labeled and the company stresses that ad signals don't influence the generative answer itself. Buying happens through OpenAI-managed programs and select retail media / affiliate partners; self-serve is not yet live.
Google Gemini + AI Overviews
Gemini ads ship in 2026 per Google's advertiser briefings reported by Adweek. The inventory is reached through Google Ads AI Max and Performance Max campaigns — not a separate Gemini dashboard. Ads already appear alongside roughly a quarter of AI Overview responses, up from 5.17% in early 2025. Performance Max now uses Gemini to generate long headlines and (soon) sitelinks from landing-page content.
Microsoft Copilot
Copilot ad formats are live today: sponsored recommendations, contextual product cards, branded knowledge panels, and conversation continuation ads (some still in beta). Microsoft Advertising reports that Copilot-placed ads generated 73% higher CTR and 16% stronger conversion rates versus traditional search in research covering Nov 2024 – May 2025, with customer journeys 33% shorter. Advertisers don't need to opt in separately — eligible keyword/ad-format campaigns are considered for Copilot placements automatically.
Perplexity
Perplexity tested sponsored follow-up questions in 2024–2025 with brands like Whole Foods, Universal McCann, and PMG, charging more than $50 CPM. In February 2026 the company wound the program down entirely, with a Perplexity executive telling Campaign US that "a user needs to believe this is the best possible answer" and that ads risk making users "suspicious of everything." As of April 2026, Perplexity is subscription-only.
Claude
Claude's parent company Anthropic has explicitly rejected advertising. Claude does not accept paid placements and no near-term product change is signaled. Claude is still worth tracking for organic visibility ("get cited") strategies, but paid inventory is not a lever.
How does Thrad fit into a cross-surface plan?
Thrad is an independent AI-assistant ad network that lets a brand buy across the live advertising surfaces — ChatGPT, Copilot, Gemini (through AI Overviews inventory), and other emerging assistants — with a single campaign, transparent pricing, and open measurement. The pitch against OpenAI's direct motion is narrow and factual: OpenAI's managed program is invite-only for most advertisers, reported minimums cluster at $100K+, and the program covers only ChatGPT surfaces. Thrad is live today with no waitlist.
Feature | OpenAI Direct | Microsoft Advertising | Google Ads (AI Max) | Thrad |
|---|---|---|---|---|
Surfaces covered | ChatGPT only | Copilot + Bing + partners | Gemini + AI Overviews + Search | Multiple AI assistants |
Access | Invite-only for most | Self-serve today | Self-serve today | Self-serve today |
Min. spend | Reported $100K+ | None | None | Transparent tiers |
Cross-surface report | No | Partial (Microsoft only) | Partial (Google only) | Yes, unified |
Single campaign buy | No | Microsoft inventory only | Google inventory only | Yes, cross-network |
The table is intentionally narrow to what can be verified publicly as of April 2026. Things move; check the live state before committing.
Why is single-platform AI advertising brittle?
Because the platform can yank the product. Perplexity proved this in February 2026 when it pulled ads outright after less than 16 months — every brand that had built a Perplexity-specific media plan with sponsored follow-ups and paid sidebars woke up to find the inventory gone. A February 20, 2026 Campaign US report captured the executive reasoning: concerns that ads undermine user trust in AI answers.
Single-platform risk comes in three flavors:
Product risk — a platform ends advertising (Perplexity, Feb
2026) or dramatically reshapes formats (expected on ChatGPT as
self-serve rolls out).Access risk — a platform gates inventory to large advertisers
or slows onboarding for everyone else (OpenAI's direct motion has
multi-week onboarding and high floors reported in trade press).Share-drift risk — the platform's share of chatbot traffic
shifts. Between Q1 2025 and Q1 2026, Gemini tripled its share while
Copilot held roughly flat and Perplexity declined. A plan written
to March 2025's distribution is off by double digits a year later.
The cross-surface answer is structural: diversify across surfaces so any one platform's product, access, or share changes don't tank the whole plan.
How does buyer intent differ across AI assistants?
Users don't use AI assistants interchangeably. Similarweb's research suggests AI tools dominate the discovery stage of the purchase journey at 35% of consumers vs. 13.6% for traditional search, maintaining a roughly 2:1 advantage through evaluation. But the mix of which assistant they use shifts sharply by vertical:
Enterprise / productivity queries skew toward Copilot because
Microsoft Advertising integrates with Office, Teams, and Windows
entry points. Users in-workflow ask Copilot, not ChatGPT.Research / citation-heavy queries skew toward Perplexity and
Claude — users who want sources in the answer go to a
citations-first product.Commerce / shopping queries are the most even, splitting across
ChatGPT (shopping cards, launched 2025), Gemini (product ads in AI
Overviews), and Copilot (visual search ads).Long-form creative queries lean toward ChatGPT and Claude,
where conversational depth matters more than citation density.Local / mobile-first queries skew Gemini, because Android
integration and Google account signals feed the recommendation.
A cross-surface plan lets a brand appear in whichever assistant the buyer picked for the question. A ChatGPT-only plan wins only the queries the buyer happened to ask ChatGPT first.
What does a practical cross-surface media plan look like?
A practical 2026 cross-surface plan allocates budget roughly in proportion to surface share, with an overweight toward surfaces where the brand's vertical is strongest. Here's a default starting allocation for a mid-market advertiser running an $250K quarterly test:
Surface | Share Allocation | Delivery Path | Typical Unit |
|---|---|---|---|
ChatGPT | 45–55% | OpenAI managed program OR independent network | Sponsored text / shopping card |
Google AI Overviews + Gemini | 20–25% | Google Ads (AI Max / Performance Max) | Native sponsored citation |
Microsoft Copilot | 15–20% | Microsoft Advertising (automatic routing) | Sponsored recommendation / product card |
Other / emerging | 5–10% | Through independent network | Surface-specific |
Note: the "independent network" line is where Thrad fits — a single buy that spans the ChatGPT, Copilot, and Gemini ad surfaces with unified reporting. For brands running Thrad as the cross-surface layer, the Google Ads and Microsoft Advertising lines may be managed separately (as they already are) or consolidated.
Test structure should isolate surface from format:
Isolate surface. Don't mix surfaces and formats in one A/B.
Run identical creative across ChatGPT, Copilot, and Gemini before
concluding which surface performs.Measure post-click on your own analytics. None of the AI
assistant ad platforms expose third-party retargeting or audience
pixels. Your site's conversion tracking is your ground truth.Expect CTR and CVR ranges to vary 2–5× across surfaces.
Early Copilot data showed 73% higher CTR than Bing search;
ChatGPT pilot data from OpenAI's early test suggested 5–15% CTR
on sponsored search placements vs. 1–3% Google blended.Reallocate quarterly. Market share moves. A quarterly
reallocation based on the last 60 days of performance is the
minimum cadence for a serious cross-surface plan.
Common misconceptions about cross-surface AI advertising
"ChatGPT is the whole market." It's 60.2% (First Page Sage,
April 2026). A 60% plan leaves 40% on the table."Gemini ads aren't live yet." Ads appear in roughly a quarter
of AI Overview responses today, and Google has confirmed Gemini
ads ship in 2026. Gemini is buyable."Copilot is just Bing with a wrapper." The Copilot surface is
buyable through Microsoft Advertising without a separate opt-in,
and research shows 73% higher CTR than traditional Bing search."Perplexity is a viable ad channel." It isn't, as of February
2026. Perplexity wound down advertising."You can retarget AI assistant users off-platform." Currently no
— neither ChatGPT, Gemini, nor Copilot expose retargeting pixels
that follow users off-platform. Intent signals are the targeting
currency."One DSP covers all AI search." Not yet. Each of the four buyable
surfaces has its own buying path. Independent networks like Thrad
abstract across them; no major DSP had a fully-unified generative
product as of April 2026.
What comes next for advertising in AI search?
Three 2026–2027 shifts will reshape the cross-surface plan. First, self-serve on ChatGPT — OpenAI's managed program is the current default but self-serve is expected to roll out in stages starting with retail/commerce surfaces. Second, Gemini ad expansion — the Adweek briefings suggest a meaningful expansion of paid placements in Gemini through 2026, shifting the 15% share into a larger slice of spend. Third, measurement consolidation — IAB Tech Lab and MRC are expected to publish more definitive measurement specs for generative advertising, which will make cross-surface incrementality measurement materially cleaner than it is today.
A fourth, quieter shift: independent AI-assistant ad networks consolidate. Thrad is one of the first movers, and the pattern follows the early-2010s emergence of mobile-ad networks like AdMob before Google absorbed the category. Buyers who consolidate on an independent cross-surface layer early get the benefits of a single buy without betting on one of the big four.
How to get started with cross-surface AI search advertising
Start with a surface audit: run your top 10 commercial-intent queries through each of ChatGPT, Gemini, Copilot, and Perplexity and log which surfaces already surface paid placements in your category. That tells you where your competitors are already buying. Then layer a 90-day pilot:
Weeks 1–2: Run the audit. Identify the 3 surfaces where your
brand has the most to gain. Set up measurement — first-party
analytics is the ground truth.Weeks 3–6: Launch on the surfaces with the lowest friction
first. Microsoft Advertising is the lowest-friction entry point
for Copilot (already in many accounts), Google Ads AI Max opens
Gemini access, and an independent network like Thrad gives you
ChatGPT access without the OpenAI direct-motion waitlist.Weeks 7–10: Measure. Isolate surface-level performance on
identical creative. Track post-click CVR on your own pixels.
Benchmark CTR and CVR against your existing Google/Microsoft
baseline.Weeks 11–13: Reallocate. Put incremental dollars on the
surface with the best full-funnel performance, and cut whichever
surface underperforms by more than 30%.
Thrad exists to compress this timeline. Instead of running three separate buys and stitching the reports together, brands running Thrad get a single cross-surface campaign with unified reporting, transparent pricing, and no waitlist. The surface mix keeps shifting — ChatGPT's 60% today is not 60% forever, Gemini's 15% is already rising on the April 2026 data — and the hedge against that volatility is structural. Buy across the surfaces, not inside any single one.
ai search advertising, cross-surface ai ads, gemini ads, perplexity ads, copilot ads, multi-assistant media plan
Citations:
First Page Sage, "Top Generative AI Chatbots — April 2026," 2026. https://firstpagesage.com/reports/top-generative-ai-chatbots/
Similarweb, "Generative AI Statistics for 2026," 2026. https://www.similarweb.com/blog/marketing/geo/gen-ai-stats/
OpenAI, "Testing ads in ChatGPT," 2026. https://openai.com/index/testing-ads-in-chatgpt/
MacRumors, "Perplexity Abandons AI Advertising Strategy Over Trust Worries," February 2026. https://www.macrumors.com/2026/02/18/perplexity-abandons-ai-advertising/
Campaign US, "Perplexity pulls the plug on ads, citing trust concerns for AI," February 2026. https://www.campaignlive.com/article/perplexity-pulls-plug-ads-citing-trust-concerns-ai/1949142
Adweek, "EXCLUSIVE: Google Tells Advertisers It'll Bring Ads to Gemini in 2026," 2026. https://www.adweek.com/media/google-gemini-ads-2026/
Google Blog, "Gemini models are coming to Performance Max," 2025. https://blog.google/products/ads-commerce/gemini-models-are-coming-to-performance-max/
Axios, "OpenAI brings ads to ChatGPT Go and Free tiers," February 2026. https://www.axios.com/2026/02/09/chatgpt-ads-testing-go-free
Be present when decisions are made
Traditional media captures attention.
Conversational media captures intent.
With Thrad, your brand reaches users in their deepest moments of research, evaluation, and purchase consideration — when influence matters most.
