Understanding what users really expect from artificial intelligence has become a key issue. While technology is advancing at a rapid pace, actual uses and perceptions remain diverse, sometimes diverging from the dominant technological narrative. Through a groundbreaking study conducted among more than 80,000 people across 159 countries and in 70 languages, Anthropic offers a rare and valuable snapshot of global expectations, uses, and perceptions of AI. Beyond the numbers, this analysis highlights a defining trend: artificial intelligence is perceived first and foremost as a tool for augmenting human capabilities.
\n\n\n\nThe originality of this study also lies in its methodology. The interviews were conducted using an adapted version of the Claude model, called Anthropic Interviewer, which is capable of facilitating open-ended conversations, tailoring its questions based on the responses, and subsequently organizing the collected data. This approach overcomes the limitations of traditional surveys by capturing more nuanced, contextualized, and qualitative responses. It also illustrates a significant shift: AI is no longer merely a subject of study; it is becoming a research tool in its own right.
\n\n\n\nOne of the key findings of the study concerns professional applications. Nearly 18.8% of respondents cited improved workplace performance as their primary expectation regarding AI. This trend confirms a reality already observed in many organizations: artificial intelligence is primarily viewed as a driver of productivity.
\n\n\n\nIn practical terms, users expect AI to handle repetitive tasks, automate certain processes, and free up time for higher-value-added activities. This trend toward increased productivity is part of a broader transformation of the workplace. According to the Boston Consulting Group, employees using generative AI tools can improve their productivity by up to 40% on certain cognitive tasks1.
\n\n\n\nThis expectation goes beyond simply increasing efficiency. It reflects a deeper aspiration: to shift the focus of work toward more creative, analytical, or strategic tasks. AI is thus viewed as a partner, capable of supporting human expertise rather than replacing it.
\n\n\n\nBeyond the professional context, the study highlights a second, more personal dimension. Approximately 13.7% of respondents view AI as a tool for personal transformation. This category includes uses related to learning, skill development, and personal growth.
\n\n\n\nHere, artificial intelligence serves as a supportive tool. It can help users understand a concept, organize their thoughts, improve their writing, or suggest ways to improve. This educational and cognitive aspect reinforces AI’s role as a tool for accessing knowledge.
\n\n\n\nAt the same time, 13.5% of users say they use AI to better manage their daily lives, particularly by organizing their schedules or making certain decisions easier. Finally, 11.1% express high expectations regarding time savings and improved quality of life.
\n\n\n\nThese findings point to a central idea: AI is seen as a tool for overall optimization, both professionally and personally. It is part of a model of continuous support, integrated into various aspects of daily life.
\n\n\n\nAnthropic’s study goes beyond expectations; it also examines user satisfaction. In this regard, the results show that adoption is already significant, though still in its early stages.
\n\n\n\nAbout 32% of respondents report having seen an increase in productivity thanks to AI. This figure confirms the tangible impact of these technologies in current applications. However, nearly 18.9% believe that AI has not fully met their expectations, particularly due to responses deemed inaccurate or insufficiently reliable.
\n\n\n\nThis discrepancy highlights a transitional phase. The technologies are already effective, but their maturity varies depending on the context of use. This situation is typical in the early stages of technology adoption, where benefits coexist with limitations that are still being addressed.
\n\n\n\nThe study also highlights several concerns expressed by users. The most common concern relates to the reliability of responses, with 26.7% of respondents citing the risk of errors or hallucinations. This is followed by concerns about the economic impact and employment (22.3%), as well as concerns about a potential loss of human autonomy (21.9%).
\n\n\n\nThese concerns should not be interpreted as a rejection of the technology. Rather, they reflect a phase of critical engagement. Users recognize the potential of AI while identifying the conditions necessary for its responsible adoption.
\n\n\n\nIt is also worth noting that approximately 11% of respondents expressed no particular concerns. For these users, AI is viewed as a technology comparable to electricity or the Internet—a tool whose benefits outweigh the risks, provided it is properly regulated.
\n\n\n\nThis diversity of views reflects the complexity of our relationship with AI. It also underscores the importance of transparency, education, and regulation in the development of these technologies.
\n\n\n\nBeyond the numbers, this study highlights a gradual convergence between technological capabilities and user expectations. AI is no longer viewed solely as a technical innovation, but as a tool integrated into everyday practices.
\n\n\n\nThis shift is accompanied by a change in user behavior. Users are no longer just looking for answers; they expect guidance, context-sensitive assistance, and the ability to adapt to their specific needs.
\n\n\n\nThis trend aligns with developments in AI models, which are increasingly focused on interaction, personalization, and multimodality. According to PwC, artificial intelligence could contribute $15.7 trillion to the global economy by 2030, largely through improvements in productivity and services2.
\n\n\n\nThe Anthropic study also highlights the importance of ethical considerations. Issues of reliability, transparency, and accountability are central to users’ expectations.
\n\n\n\nIn this context, the development of AI cannot be limited to a focus on performance alone. It must incorporate responsible design principles that ensure the quality of responses, data protection, and transparency regarding how the system works.
\n\n\n\nThese issues are now incorporated into regulatory frameworks, particularly in Europe with the AI Act, which aims to regulate the use of artificial intelligence while promoting innovation3.
\n\n\n\nThe goal is not to slow down the development of AI, but to create the conditions for its sustainable adoption, based on trust.
\n\n\n\nUltimately, what this study reveals is a shift in how artificial intelligence is viewed. Far from abstract concepts or futuristic visions, users express concrete expectations focused on efficiency, learning, and improving daily life.
\n\n\n\nAI thus emerges as a deeply pragmatic technology. It does not replace humans; rather, it enhances their capabilities, facilitates access to information, and enables better management of complexity.
\n\n\n\nThe question is no longer simply what AI can do, but how it can be meaningfully integrated into human activities. As technologies advance, this balance between technical performance and real-world utility will become a key factor in their adoption.
\n\n\n\nThe findings of this study highlight a profound shift in the use of artificial intelligence, which is directly linked to changes in the labor market and within organizations. On a related topic, check out our article “The Impact of AI on Employment: Decoding the Numbers and Trends”, which analyzes how the growing adoption of AI is redefining skills, professions, and user expectations in a context of rapid economic change.
\n\n\n\n1. Boston Consulting Group. (2023). AI and Productivity Gains.
https://www.bcg.com
2. PwC. (2023). Sizing the prize: What’s the real value of AI?
https://www.pwc.com
3. European Commission. (2024). AI Act Overview.
https://digital-strategy.ec.europa.eu
Understanding what users really expect from artificial intelligence has become a key issue. While technology is advancing at a rapid pace, actual uses and perceptions remain varied—and sometimes quite different…
\n","protected":false},"author":7,"featured_media":572342,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","footnotes":""},"categories":[29],"tags":[59],"class_list":{"0":"post-572319","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-etudes","8":"tag-parlonsia"},"yoast_head":"\n