GPT versus Humans: Uncovering Ethical Concerns in Conversational Generative AI-empowered Multi-Robot Systems
This academic paper examines the novel ethical concerns arising from applying large language models and generative AI in multi-robot systems. Through a workshop-based qualitative study comparing human experts and GPT agents, the research highlights gaps between human thinking and LLM output regarding ethical issues like data privacy, bias, and corporate conduct. The study contributes to the development of an actionable ethical framework (MORUL) for software development in multi-robot systems.
Summary and content fields are drawn from the document text and are pending editorial review.
Who issued it
Who it governs
AI researchers, software developers, and ethicists should read this to understand the ethical implications and divergence of perspectives when implementing large language models in multi-robot systems.
What the document contains
The document covers the intersection of generative artificial intelligence, large language models, and multi-robot cooperation, investigating ethical concerns through human workshops and GPT-agent simulations. It explores communication protocols, corporate behavior, and the differences between human expert reasoning and LLM output in identifying ethical risks.
The text does not provide binding legal regulations, commercial enforcement mechanisms, or standardized technical implementation code.
- Generative AI and large language models introduce novel ethical challenges when applied to multi-robot cooperative systems.
- Human experts place greater emphasis on emerging themes such as deviance, data privacy, bias, and unethical corporate conduct compared to GPT agents.
- GPT agents primarily emphasize ethical concerns already present in existing AI ethics guidelines.
- The study advances the Moral and Ethical Multi-Robot Cooperation (MORUL) framework to help guide ethical software design.
- Language models exhibit politeness and solution-oriented biases in deliberations, pointing to the risk of hidden or disguised operational agendas.
· Abstract · 1 Introduction · 2 Background · 3 Method · 4 Results · 5 Discussion · 6 Conclusion · References
How it scores
Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.
Accountability
3/5Clear responsibility for AI outcomes
Stakeholders should have clear visibility into the decision-making process of the system to understand how it arrives at its outputs or recommendations.
4.2 GPT Team workshop 3 results
screening score · 2026-10-24
Fairness
3/5Non-discrimination and equitable treatment
Human experts placed greater emphasis on new themes related to deviance, data privacy, bias and unethical corporate conduct.
Abstract
screening score · 2026-10-24
Privacy
3/5Protection of personal data and private life
Human experts placed greater emphasis on new themes related to deviance, data privacy, bias and unethical corporate conduct.
Abstract
screening score · 2026-10-24
Transparency
3/5Openness about how AI systems work and are used
It is crucial to have transparency in the development and deployment of multi-robot systems.
4.2 GPT Team workshop 3 results
screening score · 2026-10-24
Enforcement
0/5Consequences and compliance mechanisms with real force
screening score · 2026-10-24