New Study Reveals Public Perception of AI as More Confident Than Human Beings
As artificial intelligence systems gain greater popularity, more individuals are turning to them for guidance on everyday topics such as spending habits and reading recommendations. Yet a recent study published in the journal “Communications Psychology” by researchers from the University of Waterloo and University College London revealed an interesting insight. The findings indicated that even when AI and humans provide identical responses, people tend to perceive the AI’s answers as more confident.
This behavior is known as the “illusion of confidence” in AI. Research demonstrates that when people cannot assess someone’s certainty directly, they rely on external cues—such as the speed of response and the ease with which a decision is made—to form an opinion about that person’s confidence level.

Preconceptions mislead trust, lack of emotional signals leads to hidden risks
Because the public often assumes AI is more competent than humans in numerous areas, this bias can easily result in flawed judgments. Experiments have shown that once people believe an AI is capable, they assume it is highly confident in any situation, even though the system may not be reliable when faced with specific challenges.
In everyday interactions, tone of voice, facial expressions, and body language serve as crucial social signals that help us determine whether to trust someone. Since most large language models lack these expressive elements, users are forced to rely on speculation. Even if the AI is uncertain and its answer might be incorrect, users may still place undue trust in it.
Scientists explore intuitive communication methods, future large models may add new features
To mitigate this risk, Professor Kollbach and his team argue that future AI development should incorporate clearer ways to communicate a system’s actual confidence level through more straightforward and varied external indicators. This approach not only offers improvements for existing generative AI systems but also helps prevent users from making mistakes due to blind trust.
Currently, the research team is planning additional studies aimed at identifying efficient, intuitive, and trustworthy methods for human-machine communication. In the future, large language models might be equipped with features that explicitly display confidence levels, enabling users to make more informed and rational decisions about whether to follow AI-generated advice.
Related article
ByteDance’s Seed launches global campus drive, offering virtual shares to win top large model talent
In the competitive landscape of large language models, securing top-tier talent remains the most critical strategic asset.On April 1st, ByteDance announced the launch of its Seed global campus recruitment initiative, part of its large model talent de
Suno to Watermark Songs Amid Legal Battles
Suno, the platform enabling users to generate AI-created music, has unveiled new features to label platform-produced tracks, restrict downloads, and update community standards to curb unauthorized replicas. These updates arrive as Suno confronts mult
Musk Admits Grok Build Leaked User Code, Promises to Erase All Historical Data
Elon Musk directly addressed the privacy controversy surrounding Grok Build, beginning with a simple "True" to confirm the incident's validity. He pledged that all user data previously uploaded to SpaceXAI would be permanently erased, stating, "not a
Related Special Topic Recommendations
Comments (0)
0/500
As artificial intelligence systems gain greater popularity, more individuals are turning to them for guidance on everyday topics such as spending habits and reading recommendations. Yet a recent study published in the journal “Communications Psychology” by researchers from the University of Waterloo and University College London revealed an interesting insight. The findings indicated that even when AI and humans provide identical responses, people tend to perceive the AI’s answers as more confident.
This behavior is known as the “illusion of confidence” in AI. Research demonstrates that when people cannot assess someone’s certainty directly, they rely on external cues—such as the speed of response and the ease with which a decision is made—to form an opinion about that person’s confidence level.

Preconceptions mislead trust, lack of emotional signals leads to hidden risks
Because the public often assumes AI is more competent than humans in numerous areas, this bias can easily result in flawed judgments. Experiments have shown that once people believe an AI is capable, they assume it is highly confident in any situation, even though the system may not be reliable when faced with specific challenges.
In everyday interactions, tone of voice, facial expressions, and body language serve as crucial social signals that help us determine whether to trust someone. Since most large language models lack these expressive elements, users are forced to rely on speculation. Even if the AI is uncertain and its answer might be incorrect, users may still place undue trust in it.
Scientists explore intuitive communication methods, future large models may add new features
To mitigate this risk, Professor Kollbach and his team argue that future AI development should incorporate clearer ways to communicate a system’s actual confidence level through more straightforward and varied external indicators. This approach not only offers improvements for existing generative AI systems but also helps prevent users from making mistakes due to blind trust.
Currently, the research team is planning additional studies aimed at identifying efficient, intuitive, and trustworthy methods for human-machine communication. In the future, large language models might be equipped with features that explicitly display confidence levels, enabling users to make more informed and rational decisions about whether to follow AI-generated advice.
ByteDance’s Seed launches global campus drive, offering virtual shares to win top large model talent
In the competitive landscape of large language models, securing top-tier talent remains the most critical strategic asset.On April 1st, ByteDance announced the launch of its Seed global campus recruitment initiative, part of its large model talent de
Suno to Watermark Songs Amid Legal Battles
Suno, the platform enabling users to generate AI-created music, has unveiled new features to label platform-produced tracks, restrict downloads, and update community standards to curb unauthorized replicas. These updates arrive as Suno confronts mult
Musk Admits Grok Build Leaked User Code, Promises to Erase All Historical Data
Elon Musk directly addressed the privacy controversy surrounding Grok Build, beginning with a simple "True" to confirm the incident's validity. He pledged that all user data previously uploaded to SpaceXAI would be permanently erased, stating, "not a





Home






