How Americans Rely on AI Chatbots: New Data on Usage Habits and Trends
Key Highlights
- 65% rely on AI for writing and editing emails, messages, and texts, while 54% use it for brainstorming ideas.
- Users report experiencing occasional AI inaccuracies, including irrelevant responses (39%), outdated information (36%), contradictory answers (33%), and fabricated citations or sources (19%).
- Privacy concerns remain strong—54% avoid sharing sensitive personal information and 42% refrain from uploading confidential files.
ChatOn—an AI chatbot with more than 90 million downloads and powered by multiple large language models (LLMs)—has released a new survey highlighting how Americans are using AI chatbots, their level of expertise, and what concerns remain around accuracy and privacy.
The results show that AI chatbots have firmly entered everyday life. The majority of respondents use them to get answers or search for information (74%), while many rely on AI for writing and editing emails, messages, and texts (65%). More than half (54%) also turn to chatbots for brainstorming ideas.
According to the survey, 22% of Americans use AI chatbots multiple times per day, 14% use them one per day, 36% several times per week, and 16% a few times per month.
Americans are also gaining confidence in their AI skills. Nearly half (49%) consider themselves intermediate users, with 24% identifying as advanced, and 3% as experts. Meanwhile, 20% say they are casual users and 4% describe themselves as beginners.
When it comes to developing their AI proficiency, 46% prefer to experiment with prompts on their own, 27% seek tips from friends, coworkers, or classmates, and 21% learn through online tutorials such as YouTube videos. Still, 29% say they simply use chatbots without actively trying to improve.
The study also reveals the most common habits people adopt when interacting with AI tools. These include:
- Asking follow-up questions to confirm answers (48%).
- Rephrasing prompts to improve results (42%).
- Verifying AI-generated content with Google or other trusted sources (39%).
Privacy and data protection remain as significant concerns. More than half of respondents (54%) avoid sharing sensitive personal information with AI systems, 42% avoid uploading confidential files, and 36% steer clear of discussing work-related data.
Participants also reported experiencing occasional “AI hallucinations.” The most commonly noted issues—typically occurring “sometimes”—include irrelevant responses (39%), outdated information (36%), contradictory answers (33%), and fabricated citations or sources (19%).
"While most users rate their AI proficiency as intermediate or higher, their strong interest in improving their skills shows that familiarity doesn't equal mastery," said Dmitry Khritankov, Product Director, ChatOn.
Source: AIBY
Stay Connected with ISE Magazine
Subscribe to our newsletters and magazine for the latest telecom insights, explore the current issue for in-depth features and strategies, and register for upcoming webinars to learn directly from industry leaders.