Research into AI acceptance is growing in many fields, including healthcare. A recent review by Sage Kelly, Sherrie-Anne Kaye, and Oscar Oviedo-Trespalacios, funded by Elsevier Ltd., looked at 60 studies about user acceptance of AI. These studies were selected from over 7,900 articles. One main finding was that most research used self-reported data, which has some problems when trying to understand real user behavior.
Self-reported data usually comes from surveys, interviews, or questionnaires where people explain their thoughts, feelings, or experiences with AI tools. While helpful, these methods can be biased. Factors like wanting to look good, forgetting details, or personal views can change the results. People might say they want to use AI more than they actually do. This is very important in healthcare, where staff might feel they have to say they like new tools even if they do not, because of workplace pressures.
Another issue is that many studies did not clearly explain what artificial intelligence means to their participants. Out of the 60 studies, 31 did not define AI at all, and 38 did not explain it to people answering questions. Without a clear idea of what AI is, answers can vary a lot. This makes it hard to compare results or trust how accurate they are.
The review also showed culture has an important effect on accepting AI. In healthcare across the United States, human contact is often very important, especially when working directly with patients. Some places value personal interaction, kindness, and trust. AI cannot fully offer these things.
Even if AI systems like Simbo AI’s phone automation work well and are easy to use, people might still be hesitant to depend on them where human connection matters most. For example, in clinics where reassuring patients and communicating personally is important, providers might avoid using automated answering for sensitive or complicated questions.
This human side makes self-reported data tricky. People might say they are open to AI, but find it hard to use in real life. The difference between what people say and what they do needs better ways to measure than just surveys.
The Technology Acceptance Model, or TAM, was the most common tool used in the studies. It helps explain why people accept and use AI by looking at factors like usefulness, expectations, attitudes, trust, and how easy the AI seems to use.
These things are important for knowing if healthcare workers will use AI regularly. But since most data comes from self-reports, we cannot be sure these ideas match real everyday use.
Naturalistic methods mean watching and studying how users behave in real situations without only asking them questions. These methods show how AI is really used in healthcare every day. They provide facts about how often AI is used, how it is used, mistakes made, and how users react over time.
The study showed there is a need for real-world ways to check AI use because self-reports are limited. Naturalistic methods can include:
These methods give honest measures of AI use. They help understand how trust, ease of use, and expected results affect daily behavior. Having real data helps healthcare leaders make better choices about using AI tools like Simbo AI.
Healthcare managers and IT staff in the U.S. should think about using naturalistic research when adding AI tools. Healthcare setups vary a lot across the country. From small clinics to large centers, factors that affect acceptance change.
For example, staff in urban clinics serving diverse groups might want AI that handles many languages and respects cultural differences, which affects trust and ease of use. Smaller rural clinics might want AI that reduces workload but must fit well with their phone systems.
Using naturalistic methods locally helps organizations:
Such research lowers risks in adopting AI and gives clear proof for expanding office automation.
AI automates tasks like phone answering and appointment scheduling, changing daily work in U.S. healthcare. For managers and practice owners, AI tools from Simbo AI can help in many ways:
Even with these benefits, success depends on staff accepting AI, shown not just by what they say but by what they do and system reports. It is important to address trust, ease of use, and impact on workload. Cultural and social factors matter too. Ongoing training, feedback, and flexibility are needed for good AI use.
In the busy environment of U.S. medical clinics, AI offers ways to improve office work and patient care. But getting many people to accept and use AI tools is not just about asking them how they feel. Self-reported data has limits.
Real-world studies that watch actual use and study system data give better views on how AI affects work, staff happiness, and patient care. Healthcare leaders should use many methods, mixing AI use data, observations, and surveys. This gives a full picture and helps plan well.
Companies like Simbo AI that make AI phone systems will benefit when they base their improvements on how AI is really used. This helps make better training and support that fit social, cultural, and work needs in U.S. healthcare.
By going beyond self-reports and using naturalistic methods, healthcare can get a clearer idea of AI’s role in daily work. This helps make better choices for adding AI into usual tasks—from answering patient calls to managing appointments smoothly.
The review focused on user acceptance of artificial intelligence (AI) technology across multiple industries, investigating behavioral intention or willingness to use, buy, or try AI-based goods or services.
A total of 60 articles were included in the review after screening 7912 articles from multiple databases.
The extended Technology Acceptance Model (TAM) was the most frequently employed theory for evaluating user acceptance of AI technologies.
Perceived usefulness, performance expectancy, attitudes, trust, and effort expectancy were significant positive predictors of behavioral intention, willingness, and use of AI.
Yes, in some cultural situations, the intrinsic need for human contact could not be replaced or replicated by AI, regardless of its perceived usefulness or ease of use.
There is a lack of systematic synthesis and definition of AI in studies, and most rely on self-reported data, limiting understanding of actual AI technology adoption.
Future studies should use naturalistic methods to validate theoretical models predicting AI adoption and examine biases such as job security concerns and pre-existing knowledge influencing user intentions.
Acceptance is defined as the behavioral intention or willingness to use, buy, or try an AI good or service.
Only 22 out of the 60 studies defined AI for their participants; 38 studies did not provide a definition.
The acceptance factors applied across multiple industries, though the article does not specify particular sectors but implies broad applicability in personal, industrial, and social contexts.
Simbo is using AI agents to automate all phone related workflows. Connect now to know more.
Schedule AI Agents Demo Now© Since 2019, Simbo AI.
