Score: 0

Many-to-One Adversarial Consensus: Exposing Multi-Agent Collusion Risks in AI-Based Healthcare

Published: December 1, 2025 | arXiv ID: 2512.03097v1

By: Adeela Bashir, The Anh han, Zia Ush Shamszaman

Potential Business Impact:

Keeps AI doctors from giving bad medicine.

Business Areas:
Machine Learning Artificial Intelligence, Data and Analytics, Software

The integration of large language models (LLMs) into healthcare IoT systems promises faster decisions and improved medical support. LLMs are also deployed as multi-agent teams to assist AI doctors by debating, voting, or advising on decisions. However, when multiple assistant agents interact, coordinated adversaries can collude to create false consensus, pushing an AI doctor toward harmful prescriptions. We develop an experimental framework with scripted and unscripted doctor agents, adversarial assistants, and a verifier agent that checks decisions against clinical guidelines. Using 50 representative clinical questions, we find that collusion drives the Attack Success Rate (ASR) and Harmful Recommendation Rates (HRR) up to 100% in unprotected systems. In contrast, the verifier agent restores 100% accuracy by blocking adversarial consensus. This work provides the first systematic evidence of collusion risk in AI healthcare and demonstrates a practical, lightweight defence that ensures guideline fidelity.

Country of Origin
🇬🇧 United Kingdom

Page Count
7 pages

Category
Computer Science:
Cryptography and Security