Trust, but don’t verify: Americans rarely check sources amid 2.5B daily ChatGPT queries

ChatGPT now processes approximately 2.5 billion daily queries, with 330 million originating from US users, marking its explosive growth since launching in late 2022. Despite this widespread adoption, new survey data reveals a striking paradox: only 8.5% of Americans “always trust” AI-generated information from tools like Google’s AI Overviews, yet over 40% rarely verify AI responses by clicking source links.

The big picture: Generative AI has fundamentally altered how people search for information online, with ChatGPT becoming the world’s most downloaded app in April and surpassing TikTok, Facebook, Instagram, and X combined in June App Store downloads.

Key usage statistics: ChatGPT’s growth trajectory demonstrates AI’s rapid integration into daily internet use.

• The platform handles roughly 912.5 billion queries annually, though it remains far behind Google’s 5 trillion annual searches.
• ChatGPT could accelerate its growth once OpenAI launches its anticipated web browser.
• Google has responded by pivoting to AI Overviews and launching AI Mode in May, which operates similarly to competitor Perplexity, an AI-powered search startup.

Trust versus behavior disconnect: Survey data from over 1,100 Americans reveals a troubling gap between AI usage patterns and user confidence.

• Only 8.5% of respondents said they “always trust” information from Google’s AI Overviews.
• Around 21% reported having zero trust in the feature’s accuracy.
• More than 40% rarely or never click on accompanying source links to verify AI-generated responses.

What affects AI trustworthiness: Research shows that user trust in AI systems varies significantly based on context and communication style.

• A separate study found most people trust ChatGPT more than human experts in at least one domain.
• Trust diminishes when AI discusses sensitive topics like legal or medical advice.
• More sycophantic AI responses tend to be perceived as less trustworthy compared to neutral, less flattering tones.
• Large language models often “hallucinate” by providing inaccurate information with convincing presentation.

Industry response: Leading AI developers are investing in transparency initiatives to address trust concerns.

• Companies like OpenAI and Anthropic have launched programs to interpret and understand AI system processes.
• These efforts aim both to improve system performance and boost user confidence.
• The initiatives reflect growing recognition that technical capability alone isn’t sufficient for widespread AI adoption.