AI Hallucinations Concern Users More Than Job Displacement, Survey Finds

• March 23, 2026

Contrary to earlier assumptions that artificial intelligence (AI) primarily threatens employment for certain professions, new data indicates that users are more troubled by AI’s propensity to produce errors and hallucinations. This insight emerges from a survey conducted among users of Anthropic’s AI assistant, Claude.

Concerns Over AI Accuracy Surpass Employment Fears

Since the rise of AI technologies, widespread anxiety has centered on the potential for machines to replace human workers, especially in roles that involve routine or repetitive tasks. Although these concerns remain relevant, the survey findings suggest that end users now harbor deeper apprehension about the reliability and trustworthiness of AI outputs.

Hallucinations—instances where AI generates false, misleading, or fabricated information—are a significant point of worry. These inaccuracies can affect decision-making processes and diminish confidence in AI systems, leading to hesitance in integrating such technologies into critical workflows.

The survey focused on Anthropic Claude, a conversational AI known for its advanced language understanding capabilities. By gauging users’ attitudes toward Claude’s performance, researchers aimed to identify the primary sources of AI-related unease. The dominant sentiment expressed involved the uneasy experience of encountering flawed or hallucinated content rather than fears of job market disruption.

This shift in concerns highlights a growing recognition of the challenges in ensuring AI safety and accuracy. Developers and companies investing in AI technologies may need to prioritize reducing hallucinations to build greater user trust and increase adoption across professional and consumer settings.

While employment impacts remain part of the broader discourse on AI’s societal effects, the immediate user experience appears increasingly defined by the need for reliable and factually correct AI assistance. Improving this aspect could enhance the perceived value and safety of AI systems, which continues to be a significant consideration for users engaging with conversational AI platforms.

As the AI landscape evolves, understanding user priorities and fears will be crucial for shaping future developments. This survey underscores the importance of addressing the technical challenges that contribute to hallucinations, alongside ongoing conversations about workforce automation and economic implications.

A recent survey reveals that fears over AI generating errors and hallucinations outweigh concerns about job losses among users of Anthropic Claude.