Has ChatGPT actually replaced Google and sites like Stack Overflow, or just joined them as another option?
Does ChatGPT displace search engines or question-and-answer platforms?
This explores whether people have actually stopped using Google and sites like Stack Overflow since ChatGPT arrived, or whether it sits alongside them, and what changes for the people and platforms involved either way.
This explores whether ChatGPT has replaced traditional search and Q&A sites, or only added another option next to them. The collection gives a split answer: for public Q&A communities the evidence points to real displacement, while for everyday web search it points to coexistence.
The clearest case of displacement is Stack Overflow. One study compared it with four Q&A sites where ChatGPT was less useful, and found weekly posts on Stack Overflow fell 16% after ChatGPT launched, reaching about 25% within six months Did ChatGPT cause Stack Overflow posting to decline?. The drop was largest for programming languages that are well represented in ChatGPT's training data. That data came partly from public Q&A sites like Stack Overflow, so the questions it answers best are the ones those communities had already answered in public. You might expect ChatGPT to take only the easy duplicate questions. Vote scores on the remaining posts barely changed, which suggests it also took good questions Did ChatGPT displace only low-quality Stack Overflow posts?. One caveat: votes are only a rough stand-in for quality, and no experts checked the posts.
Search looks different. In Nielsen Norman Group's study, every participant kept using traditional search, and many ran it alongside AI chat on the same task Does generative AI chat actually replace traditional search?. The main limit on AI use wasn't distrust. People weren't sure when chat was the right tool. Usage data supports this. Most ChatGPT use is now outside work (over 70%), and work use centers on writing rather than looking things up How is ChatGPT actually being used by real people?. Much of what ChatGPT does wasn't something people used search for in the first place.
Where ChatGPT does replace looking things up, two costs show up. In an 8-day field experiment, people who learned with ChatGPT scored lower than Google searchers, especially on critical-thinking questions Does ChatGPT harm informal learning compared to Google Search?. The researchers say this is because ChatGPT chooses the information for you, leans toward ready-made answers rather than the principles behind them, and its conversational format means you explore less of the topic. The second cost is losing the source. In a study of more than 3,000 AI answers about news, nearly half had significant errors and a third had sourcing problems Do AI assistants reliably answer questions about news?. In a study of election questions, chatbots made no factual errors by early 2026 but sent voters to official state websites less than half the time Do AI chatbots give voters accurate election information?. Even an accurate answer can leave you without a source you can check.
The question also runs the other way. AI is changing the web that search engines index. AI-written articles grew to about half of new web articles and have stayed near 50% since early 2025 Has AI-generated content stopped growing on the web?. One possible reason for the plateau is that AI-written articles don't rank well in search, which would mean search engines still shape what gets written. Overall, ChatGPT has taken significant traffic from the public Q&A communities it learned from, while search has held on as a check on it. This collection doesn't yet have long-term data on whether search usage itself has declined.
Sources 8 notes
A difference-in-differences model comparing Stack Overflow to four Q&A sites with lower ChatGPT exposure found weekly posts dropped 16% post-launch, growing to around 25% within six months. The effect was largest for programming languages with more training data in ChatGPT.
Vote scores on Stack Overflow showed no significant change after ChatGPT's release, suggesting the displaced content included high-quality posts, not merely duplicates or poor-quality material. However, this conclusion relies on votes as a proxy for quality without expert validation.
Nielsen Norman Group's qualitative study found all participants continued using traditional search throughout tasks, often running both methods in tandem. The main barrier to AI adoption is not resistance but lack of awareness about when and how to use AI chat for information-seeking.
Analysis of representative ChatGPT conversations shows non-work use growing from 53% to over 70%, while work use centers on writing tasks rather than information retrieval, concentrating among educated professionals.
In an 8-day field experiment, ChatGPT users scored lower on knowledge tests, especially on critical thinking items, due to reduced agency in information selection and two distortions: ChatGPT's bias toward solutions over principled knowledge, and its conversational interface reducing exploration of the broader knowledge space.
Show all 8 sources
A coordinated study of 22 public broadcasters in 18 countries had journalists evaluate over 3,000 AI responses on news topics. Nearly half contained significant errors, a third had serious sourcing problems, and a fifth showed major accuracy issues like hallucinations.
States United found ChatGPT and Google AI reached 0% verifiable factual error rates by early 2026, yet directed voters to official state election websites less than 50% of the time. Incomplete candidate information and reliance on editable sources like Wikipedia further limited utility.
Graphite's analysis of 55.4k web articles found AI-primary content rose to 35.9% within 12 months of ChatGPT, reached 48% by month 24, then stabilized near 50% from Q1 2025 onward. The plateau may reflect poor search performance of AI articles.
Papers this line draws on 8
The research behind the notes this line reads — ranked by how closely each paper relates.
- Are Large Language Models a Threat to Digital Public Goods? Evidence from Activity on Stack Overflow
- How AI Is Changing Search Behaviors
- AI and Elections: How Well Do AI Platforms Answer Voter Questions?
- Emerging uses of AI chatbots for news and what it means for journalism (Digital News Report 2026)
- Digital News Report 2026
- The effect of ChatGPT on students’ learning performance, learning perception, and higher-order thinking: insights from a meta-analysis
- Strengthening ChatGPT's responses in sensitive conversations
- Learning by Chatting? Investigating the Impact of Generative AI on Information Seeking and Learning