AI search statistics
From Search Visibility Wiki · By the Search Visibility Wiki editorial team · Published · Last reviewed · How we research
These AI search statistics are taken only from primary sources, with dates. Highlights: OpenAI reported 700 million weekly ChatGPT users (September 2025); Pew found clicks on 8% of Google visits with an AI summary vs 15% without; SparkToro found 58.5% of US Google searches ended without a click (2024).
Usage
- More than 700 million people use ChatGPT each week (OpenAI, September 29, 2025).[1]
- 58% of Pew’s 900-adult US panel saw at least one Google AI summary in March 2025.[2]
Clicks
- Users clicked a traditional result in 8% of visits with an AI summary vs 15% without; clicks inside the summary in 1% (Pew, March 2025 data).[2]
- AI Overviews correlated with 34.5% lower position-1 CTR (Ahrefs, 300,000 keywords, April 2025).[3]
- 58.5% of US and 59.7% of EU Google searches were zero-click (SparkToro, 2024).[4]
Accuracy and research
- 51.5% of sentences fully supported by citations; 74.5% of citations supportive (Liu et al., 2023).[5]
- GEO methods raised visibility by up to 40% in the original benchmark.[6]
- Branded web mentions correlated 0.664 with AI Overview brand visibility (Ahrefs, 75,000 brands).[7]
Frequently asked questions
Why so few statistics?
We include only figures we could verify at the primary source.
How often is this updated?
When sources publish new data.
Can I cite this page?
Yes; cite the primary source too.
See also
References
Sources accessed October 14, 2026 unless another date is given. See all sources.
- “Buy it in ChatGPT: Instant Checkout and the Agentic Commerce Protocol”. OpenAI. September 29, 2025.
- “Google users are less likely to click on links when an AI summary appears in the results”. Pew Research Center. July 22, 2025; March 2025 browsing data.
- “AI Overviews Reduce Clicks by 34.5%”. Ahrefs. April 17, 2025.
- “2024 Zero-Click Search Study”. SparkToro (data from Datos). July 1, 2024.
- “Evaluating Verifiability in Generative Search Engines”. arXiv (Liu, Zhang, Liang; Stanford). April 2023; Findings of EMNLP 2023.
- “GEO: Generative Engine Optimization”. arXiv (Aggarwal, Murahari, Rajpurohit, Kalyan, Narasimhan, Deshpande). v1 submitted November 16, 2023; published at KDD ’24.
- “An Analysis of AI Overview Brand Visibility Factors (75K Brands Studied)”. Ahrefs. May 26, 2025.
Categories: Reference