ChatGPT Is No Longer the Only AI Search Engine That Matters. Here Is What the Data Shows.

For most of 2025 and early 2026, "AI visibility" effectively meant "ChatGPT visibility." The platform dominated referral traffic so thoroughly that optimizing for it alone felt like a reasonable strategy. That is no longer the case.

Goodie's 2026 AI Search Traffic Report, published May 21, 2026, tracked B2B AI referral data across platforms and found that ChatGPT's share dropped from 89% to 62.6% in eight months. Claude surged from 1.4% to 18.5%. Gemini quadrupled to 10.6%. Perplexity more than doubled to 7.3%. The report's framing: "The Big 1 has become the Big 4" (Goodie, May 2026).

A ChatGPT-only optimization strategy now ignores more than a third of the AI referral landscape. Here is what the data shows, where the traffic is shifting, and what it means for multi-platform AI visibility.

The Market Share Data

Multiple sources are tracking this shift, and the numbers vary depending on what is being measured. The key distinction: market share of users, share of web traffic, and share of B2B referrals are three different metrics that produce three different pictures.

B2B AI referral share (Goodie, March-April 2026)

Goodie's methodology triangulates against Similarweb and SensorTower data. The report notes that "AI platform popularity is not the same as B2B referral output. The ratio between the two varies by 10x or more across the four major platforms" (Goodie, 2026). Claude's growth is particularly notable: it went from a rounding error to the second-largest source of B2B AI referrals.

Global web traffic share (Similarweb, March 2026)

Similarweb's global AI tracker tells a broader story. ChatGPT's share of worldwide AI web traffic dropped from 86.7% in January 2025 to 56.7% by March 2026. Gemini surged from 5.3% to 25.5% over the same period. This measures overall web traffic rather than B2B referrals specifically, which is why the numbers differ from Goodie's (Similarweb, OfficeChai, March 2026).

AI chatbot market share (First Page Sage, May 2026)

First Page Sage's May 2026 AI chatbot market share report shows a different distribution again: ChatGPT at 53.1%, Claude at 20.9%, Gemini at 13.2%, Copilot at 8.7%, and Perplexity at 2.8% (First Page Sage, May 2026). These numbers measure overall chatbot usage, not search-specific behavior, which explains the lower Perplexity figure (Perplexity is primarily a search product) and the higher Claude and Copilot numbers (driven by enterprise and developer adoption).

The details vary across reports, but the trend is consistent: ChatGPT is losing share across every measurement, and three other platforms are absorbing it.

ChatGPT Still Dominates Raw Volume

The share decline does not mean ChatGPT is shrinking. ChatGPT now processes approximately 2.5 billion daily prompts. An Ahrefs analysis estimated that roughly 65% of those prompts are "search-like queries," which translates to approximately 1.6 billion daily search queries, 12% of Google's estimated 13.7 billion daily searches and more than Bing's approximately 1.2 billion (Ahrefs, Patrick Stox and Ryan Law, February 2026).

But volume and traffic are different things. Ahrefs found that across 76,000 websites, Google sends approximately 190 times more traffic to websites than ChatGPT. Google accounts for 39.98% of all website traffic versus ChatGPT's 0.21%. The CTR gap is enormous: Google's search CTR is 29.2% versus ChatGPT's 1.3%, a 96% reduction (Ahrefs, February 2026).

In other words, ChatGPT processes massive query volume but sends very little of it back to websites. This is a structural feature of conversational AI, not a temporary limitation.

The May 7 Branded Link Update Changed the Picture

On May 7, 2026, ChatGPT began embedding brand homepage URLs inline in its answers far more frequently. Similarweb tracked the impact across desktop traffic and found that total ChatGPT referrals increased 157.7% week-over-week. Homepage referrals specifically surged 354.7% (Similarweb, May 2026).

Qwairy analyzed over 140,000 ChatGPT answers and found that the share of responses containing a branded link jumped from 0.4% to 6.2% overnight, a step change rather than a gradual ramp. Among answers that mention a brand, the linked share rose from approximately 2% to 29%. 79% of the new links point to the brand's homepage, up from 59% (Qwairy, May 2026).

The behavioral data after the update was also notable: page views per visit increased from 3.8 to 4.7 (24% uplift) and time on site increased from 3.5 to 3.9 minutes (11% uplift). Every link includes utm_source=chatgpt.com tracking. Crucially, only ChatGPT implemented this change. Perplexity, Gemini, and Copilot showed no corresponding shift during the same period (Similarweb, 2026).

Each Platform Cites Differently

The Tinuiti Q1 2026 AI Citation Trends Report, developed with Profound, tracked high-commercial-intent prompts across 9 verticals and 7 AI platforms (ChatGPT, Perplexity, Google AI Mode, AI Overviews, Gemini, Copilot, Meta AI). The core finding: "There's no universal top source. There are only patterns shaped by intent, platform, and category" (Tinuiti, 2026).

Reddit's surge in AI citations

Reddit's citation share across AI platforms grew by at least 73% from October 2025 to January 2026. On Perplexity specifically, 24% of all citations in January came from Reddit alone (Tinuiti, 2026). Social media citation share overall topped 9% by January 2026, with Reddit as the dominant contributor.

The Reddit citations are not going to brand pages or subreddit homepages. Tinuiti found that 99% of Reddit citations on ChatGPT point to unique discussion threads containing useful answers. YouTube is the only other social platform to consistently exceed 1% of citations across platforms.

Platform-specific citation patterns

• ChatGPT: Cites Reddit in over 5% of responses. After the May 7 update, increasingly links to brand homepages inline. Drives 87.4% of all AI referral traffic by raw volume (TechnologyChecker, 2026).
• Perplexity: Draws 31% of its citations from social media, with Reddit dominant at 24% of all citations. The heaviest use of forum and community content.
• Gemini: Cites Reddit in just 0.1% of responses. Relies more heavily on authoritative publisher content and Google's own index.
• AI Overviews: Reddit accounted for 44% of all social media citations in January, but social citations are still a small fraction of total AI Overview sources. 96% of AI Overview links come from top-10 organic results (Brainlabs, 2026).

Citation volumes can differ by 615 times between platforms for the same query category. Grok and Claude occupy opposite ends of that spectrum. What gets cited on Perplexity has almost nothing to do with what works on Gemini (Tinuiti, 2026).

AI Traffic Converts at Higher Rates

Despite low volume (AI referral traffic accounts for approximately 1% of all website traffic), the conversion data is striking. Ahrefs reported that AI search visitors to their site converted at 23 times the rate of traditional organic search visitors, with 0.5% of traffic driving 12.1% of signups (Ahrefs, Patrick Stox and Ryan Law, June 2025). This is a single B2B SaaS company and should not be generalized as a universal rate.

Broader industry data shows a wide range. Semrush reported a cross-industry average of approximately 4 times higher conversion for AI referrals (RunMarshal, 2026). A Microsoft Clarity study across 1,277 publisher domains found Copilot referrals convert at 17 times the rate of direct traffic. Loamly's benchmark report found that "dark AI" traffic (AI referrals without referrer headers, which accounts for 70.6% of AI traffic) converts at 10.21% versus 2.46% for non-AI traffic, a 4.1 times multiplier (Loamly, February 2026).

The range is 1.3 times (low-consideration e-commerce) to 23 times (B2B SaaS). The directional finding is consistent: AI-referred visitors are higher-intent than average search visitors, likely because conversational AI filters and pre-qualifies the query before sending the user to a site.

What This Means for Multi-Platform Strategy

The data points to four practical implications for AI visibility strategy:

• A ChatGPT-only strategy now misses a third of the market. Eight months ago, optimizing for ChatGPT covered 89% of B2B AI referrals. Today it covers 63%. Claude, Gemini, and Perplexity each have different retrieval systems, citation patterns, and content preferences. Multi-platform tracking is no longer optional for brands that take AI visibility seriously.
• Reddit is now a major citation source, but only on certain platforms. Perplexity pulls 24% of its citations from Reddit. Gemini pulls 0.1%. If your brand has an active Reddit presence with helpful, detailed responses in relevant threads, that content is being cited, but only on platforms that index social content heavily. This is not a universal strategy.
• The branded link update matters for ChatGPT specifically. The May 7 change nearly tripled ChatGPT referral traffic and shifted 79% of links to homepages. For brands being mentioned in ChatGPT responses, homepage optimization for AI-referred visitors (clear value propositions, fast load times, conversion paths) is now directly tied to traffic volume. This did not happen on other platforms.
• Conversion rates justify the effort despite low volume. AI referral traffic is approximately 1% of total web traffic. But with conversion rates 4 to 23 times higher than traditional search (depending on vertical), that 1% can drive outsized business impact. The math works differently than volume-based SEO metrics.

The Bottom Line

The AI search market has fragmented. ChatGPT is still the largest platform by every measure, but it is no longer so dominant that you can ignore everything else. Claude's jump from 1.4% to 18.5% of B2B referrals happened in eight months. Gemini's web traffic share went from 5.3% to 25.5% in just over a year. These are not gradual trends.

Each platform retrieves content differently, cites different sources, and weights different signals. What ranks on Google does not automatically appear on ChatGPT. What ChatGPT cites bears little resemblance to what Perplexity surfaces. A brand that tracks only one platform is optimizing with partial data.

The practical takeaway: audit your brand's visibility across ChatGPT, Claude, Gemini, and Perplexity. Track citation patterns on each platform separately. And stop treating "AI visibility" as a single channel, because it is four channels with four different playbooks.