Anthropic said Wednesday that it has completed what it calls the largest and most multilingual qualitative study yet on how people feel about AI, based on 80,508 interviews with Claude users across 159 countries and 70 languages conducted over one week in December. The company used “Anthropic Interviewer,” a version of Claude designed to run structured, adaptive one-on-one conversations, effectively making the model itself the interview tool.
The clearest takeaway was that users are not mainly asking AI to make them work faster. Anthropic identified nine broad categories of desired outcomes, led by professional excellence at 18.8%, personal transformation at 13.7%, and life management at 13.5%. But the company said many respondents who began by talking about productivity eventually described a more personal goal: gaining time, flexibility, and a better life outside work. Anthropic also found that 81% said AI had already moved them at least somewhat toward that vision, with productivity gains and cognitive partnership among the most commonly cited benefits.
The study also found that concern is more fragmented than optimism. Respondents reported an average of 2.3 distinct worries each, with unreliability — including hallucinations, inaccuracies, and the need for constant verification — emerging as the most common at 26.7%. Economic and job-related fears followed at 22.3%, while concerns about loss of human agency reached 21.9%. Anthropic said economic anxiety was the strongest predictor of overall sentiment toward AI.
Even so, the global mood leaned positive. Anthropic said 67% of respondents expressed an overall positive view of AI, and no country in the dataset fell below 60% positivity. The company noted, however, that optimism was weaker in Western Europe and North America than in parts of South America, Africa, and Asia.
