Human Purpose in Peril: Americans fear AI will undermine human meaning and weaken social bonds, even as AI users report some satisfaction and useful outcomes from the bots now

Key Findings

• — Among LLM users, 51% said they felt lazy or took easy shortcuts at least some of the time when using bots, 39% said they are growing overly dependent on LLMs for answers rather than thinking for themselves, 39% feel they are cheating at times, and 22% said they have felt manipulated
• — 69% of US adults said AI will weaken human-to-human bonds a lot (33%) or somewhat (36%); 64% agreed that by replacing humans in most tasks AI will undermine people's sense of purpose and meaning
• — 55% are not confident that people will retain their ability to think critically and evaluate information (15% confident); 50% are not confident most people will use AI appropriately
• — 65% of US adults used an LLM in the past week, up from 52% in the Center's earlier survey
• — Users also credit bots with helping them learn, offering better advice than family or friends at times, and helping them prepare for difficult human conversations

Methodology Notes

Survey of 1,505 US adults including an oversample of 501 LLM users (1,163 users in total), 14 to 18 May 2026; margin of error plus or minus 3.1 points for the full sample. An LLM user is a respondent who reports using ChatGPT, Gemini, Copilot, Meta AI, Claude, Grok, DeepSeek or another AI chatbot. Topline and methodology released as a separate PDF. Date precision: report page dated 14 September 2026; PDF cover 'September 2026'. Curator fetched the report page and the PDF on 2026-09-17 and read the executive summary.

Sources