Searching the web used to mean scanning a list of blue links. Now many searches return a written answer. Google shows AI Overviews and, since rolling it out across the US in May 2025, offers an AI Mode, ChatGPT and Claude can search the web inside a conversation, and dedicated answer engines such as Perplexity are built entirely around AI summaries with citations.
Updated September 2026: we added Google’s AI Mode rollout and Pew Research Center data on how AI summaries affect clicks to websites.
So which should you use? It depends on what you are trying to do.
How AI search works
AI search tools run one or more web searches, read the top results and have a language model write an answer, usually with links to sources. It is a form of retrieval-augmented generation applied to the web.
Side by side
| Task | Better choice | Why |
|---|---|---|
| Quick factual questions | Either | Both usually get it right; check the source for anything important |
| Complex, multi-part questions | AI search | It can combine information from several sources |
| Comparing options | AI search, then verify | Good for a first overview |
| Finding a specific website | Traditional | Faster and more direct |
| Local businesses, maps, hours | Traditional | Richer, more current listings |
| Breaking news | Traditional or news sites | AI summaries can lag or mix old and new |
| Shopping and prices | Traditional, retailer sites | Prices change quickly |
| Research where sources matter | AI search with citations, then read sources | Summaries can misread or overstate |
Strengths of AI search
- Saves time on questions that would require opening several pages
- Handles natural, conversational questions and follow-ups
- Good at explaining concepts and summarizing perspectives
Weaknesses of AI search
- Summaries can be wrong even with good sources, by misreading, oversimplifying or blending sources together. Our hallucinations guide explains the pattern.
- Source quality varies. An answer is only as good as the pages it draws from.
- Less exploration. You see fewer perspectives than when you scan results yourself.
Strengths of traditional search
- Direct access to primary sources
- You judge source credibility yourself
- Better for navigation, local information and very recent content
A good habit Use AI search to get oriented, then click through to at least one strong primary source, such as an official site, a reputable publication or original research, before you act on anything important.
What it means for the web
Publishers have raised concerns that AI summaries reduce clicks to the sites that produce the original information, and the data supports them. A Pew Research Center study of real browsing in March 2025 found that Google users who saw an AI summary clicked a traditional result in 8% of visits, against 15% when no summary appeared. Links cited inside the summary were clicked in just 1% of visits. How search engines and AI companies share traffic and value with publishers is an active debate, and it affects what content gets made in the future.
Use both, for different jobs
AI search is excellent for understanding and synthesizing; traditional search is better for finding, navigating and verifying. Most people will get the best results by using both, and by always keeping one eye on the sources.
Sources
- AI Mode in Google Search: Updates from Google I/O 2025, Google, May 2025
- Google users are less likely to click on links when an AI summary appears in the results, Pew Research Center, July 2025
