The Illusion of Trust in AI
Behavioural Differences Between Humans and Large Language Models
Bibliographic Data
| ID | 23039697 |
|---|---|
| Authors | Yuzhan Hang (0000-0002-2970-0399, University of Science and Technology of China), Zhenhua Ling (0000-0001-7853-5273, University of Science and Technology of China), Quan Liu (0000-0002-0941-8767, Ian's Friends Foundation), Xiaosong He (0000-0002-7941-2918, University of Science and Technology of China, corresponding author), Wei Wu (0000-0003-1901-9134, University of Science and Technology of China, corresponding author) |
| Editors | Tze Wei Liew (0000-0003-0153-8649) |
| Year | 2026 |
| Volume | 2026 |
| Issue | 1 |
| Publication date | 2026-01-01 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Human Behavior and Emerging Technologies (JOURNAL) |
| Journal identifiers | ISSN: 2578-1863 • E-ISSN: 2578-1863 |
| Publisher | Wiley (PUBLISHER • GB) |
| DOI | 10.1155/hbe2/3628473 |
| OpenAlex | W7139133395 |
| Language | EN |
| References cited | 26 |
As artificial intelligence (AI) systems increasingly enter trust‐dependent domains, questions arise about whether their behaviour reflects genuine trustworthiness or merely the illusion of it. This study examined how humans and large language models (LLMs) establish and adjust trust in dynamic social interactions using a 50‐round trust game. Across 100 human participants and three leading LLMs—ChatGPT‐3.5, ChatGPT‐4o and DeepSeek‐V3—we compared trust trajectories, responsiveness to partner behaviour and reactions to unexpected outcomes. Human participants adjusted trust in line with partner trustworthiness and exhibited symmetrical responses to unexpected gains and violations. In contrast, LLMs showed fixed, model‐specific behaviour with little to no adaptation based on interaction history. Despite their cooperative appearance, AI agents lacked mechanisms for social learning and trust calibration. These findings highlight a fundamental disconnect between perceived and actual AI behaviour and underscore the need for cautious interpretation of AI trust signals in socially sensitive contexts.
Adaptation (eye) · Dictator game · Human behavior · Illusion · Interpretation (philosophy) · Social relation · Trait · Trustworthiness · AI in Service Interactions · Artificial Intelligence in Healthcare and Education · Explainable Artificial Intelligence (XAI
Cooperation and Punishment in Public Goods Experiments
Positive reciprocity and intentions in trust games
On seeing human
Not So Different After All
Trust, Reciprocity, and Social History
AI Psychometrics
The AI doctor will see you now
The cultural stereotype and cultural bias of ChatGPT
Defection in the dark? A randomized‐response investigation of cooperativeness in social dilemma games
On preferences and doing the right thing
Machines and Mindlessness
Socially desirable responding and its elusive effects on the validity of personality assessments
The Norm of Reciprocity
| Citation velocity | historical |
|---|---|
| Highly cited | No |