Objective:

To outline studies addressing the user experience of LLM-based conversational agents for physical activity promotion and to discuss the ethical implications of their persuasive design.

Approach:

• Review Methodology: A mini-review was conducted to synthesize emerging evidence from studies on LLM-based conversational agents and physical activity, focusing on user experience and social dynamics.

Key Findings:

• LLM-based conversational agents show positive effects on user engagement, but evidence for a direct, sustained impact on objectively measured physical activity remains limited.
• These agents can assume various social roles, influencing relational dynamics with users.
• Users tend to anthropomorphize LLMs, which can enhance emotional investment but may lead to over-reliance and unrealistic expectations.

Interpretation:

The review discusses the potential and challenges of LLMs in promoting physical activity, highlighting the need for a deeper understanding of user interactions and ethical concerns.

Limitations:

• The review is based on a limited number of studies, which may not provide a comprehensive overview.
• The methodology did not involve independent checks by a second reviewer for data extraction and quality appraisal.

Conclusion:

The findings indicate a need for further research into the user experience of LLM-based conversational agents and the ethical implications of their persuasive capabilities.