Consumer data shows broad resistance to AI agents handling daily tasks

Meta's Muse agent reached the top of Apple's App Store, driven by integrations with Facebook and Instagram and a fuzzy mascot design. The company has since expanded the tool to business customers, fueling industry hopes that agentic AI can break through to mainstream users. However, new polling data suggests a significant gap between this early success and the willingness of the average American to grant AI agents control over their digital lives.

The enthusiasm for these tools is not universal. A Pew survey conducted in February found that 51% of Americans avoid using AI chatbots entirely. Among those who do not use chatbots, 79% cited privacy concerns, and 67% said they were unlikely to adopt the technology in the coming year. This reluctance extends to the specific capabilities that define agentic AI. A Thales international poll found that only 13% of respondents would allow an AI helper to read their emails, while just 11% would permit an agent to rebook travel. Financial control remains a particularly sensitive boundary, with only 7% willing to let an agent move money between bank accounts.

Shopping presents another hurdle. YouGov data indicates that 56% of respondents would not allow an agent to shop for them at all. Trust in autonomous spending is low, with only 10% comfortable letting an agent spend more than $25 without prior approval. These figures suggest that the current user base for AI agents may not reflect the broader population.

Menlo Ventures, working with Morning Consult, identified a demographic skew in current usage patterns. Households with incomes over $100,000 are substantially more likely to use AI than those earning under $50,000. Users who pay for AI services are twice as likely to use it daily and five times more likely to use agents. Menlo Ventures describes its core power users as millennial parents with post-graduate degrees working in technology or financial services, noting that this group tends to have more money than time.

The St. Louis Federal Reserve describes AI adoption as "widespread but shallow," stating that at least 20% of workers use AI in more than 80% of occupations. However, this adoption is heavily concentrated in white-collar and tech-adjacent roles. Computer programmers, public relations professionals, financial advisors, and CEOs report usage rates around or above 80%. In contrast, licensed practical nurses, receptionists, and animal caretakers report usage rates at or below 10%. The Federal Reserve notes that current AI tools primarily work on screens and cannot perform physical tasks like driving trucks or drawing blood.

Internal data from AI developers reflects similar trends. Anthropic's Economic Index Survey, which the company notes is not representative of the general population, found that computer and math-related professions made up 30% of respondents despite representing only 4% of U.S. employment. Management roles accounted for 23% of respondents against a 7% share of employment.

Gallup data highlights a disconnect between AI utility and perceived life importance. Only 6% of Americans consider utilizing tools like ChatGPT a very important ingredient for a good life. This compares to 78% who value spending time with partners, friends, or family, and 58% who prioritize physical activity.