Overview

Tairex, an AI healthcare technology company incubated by the Institute for AI Industry Research (AIR) at Tsinghua University, announced on October 21, 2026 the launch of its Multidisciplinary Human‑Agent Collaboration (MHAC) Ecosystem for Agent Hospital. The press release, issued from Beijing on October 22, 2026, positions the system as a means to enable specialized AI agents to work alongside physicians on clinical cases, addressing the challenge of coordinating multiple AI outputs for a single patient.

System Architecture and Training

The MHAC Ecosystem is built on a simulated environment populated by millions of highly realistic synthetic AI patients. These patients are generated partially from clinical guidelines and medical literature rather than real medical records, reducing privacy risks. Within this virtual training ground, an AI doctor can conduct tens of thousands of simulated consultations daily. The resulting platform comprises 54 AI doctor avatars and millions of AI patients, covering more than 20 clinical specialties and tens of thousands of diseases. The AI doctors have achieved over 96% accuracy on the MedQA dataset and can adapt to physicians’ consultation styles and diagnostic reasoning through continuous clinical feedback.

Clinical Collaboration Capability

The core capability of the MHAC Ecosystem lies in organizing specialized doctor‑avatar agents into a coordinated virtual team. When a patient’s condition spans multiple specialties, the system dynamically activates the relevant agents to work simultaneously. All agents present their clinical reasoning and supporting evidence on a unified interface, with any disagreements highlighted for physician review. For example, an endocrinology agent suggested a beta‑blocker for a patient with palpitations and a history of asthma; a respiratory medicine agent flagged the risk of bronchoconstriction from non‑selective beta‑blockers and recommended a selective beta‑1 blocker, illustrating how cross‑specialty AI collaboration can surface risks that a single‑specialty tool might miss.

Impact on Clinical Workflows

Multidisciplinary team (MDT) consultations are essential for complex diseases but are often hampered by scheduling conflicts, gaps between recommendations and treatment execution, and fragmented patient journeys. The MHAC Ecosystem addresses these issues by delivering multidisciplinary input during the initial consultation, allowing physicians to consider simultaneous recommendations from multiple specialist agents without waiting for separate meetings. The system also extends specialist expertise to multiple departments and potentially to primary‑care facilities, improving access to high‑quality medical resources across a broader range of healthcare institutions.

Design Philosophy and Value Measurement

Tairex emphasizes that AI is intended to improve efficiency and support clinical decision‑making rather than to make autonomous critical decisions. The value of the AI‑assisted ecosystem is to be measured by tangible improvements in physicians’ time utilization, quality of care, patient experience, and access to primary‑care services.

Data Security and Regulatory Compliance

Data security is integral to the ecosystem’s design. Tairex’s underlying products reference the European Union’s General Data Protection Regulation (GDPR) and South Korea’s Personal Information Protection Act (PIPA), with controls governing the handling and transfer of sensitive data fields. Additionally, the system has obtained China’s Multi‑Level Protection Scheme (MLPS) Level 3 certification, underscoring its compliance with stringent data protection standards.

Outlook

By integrating AI‑assisted workflows with human‑led clinical decision‑making, Tairex aims to help high‑quality medical resources reach more healthcare institutions, including those at the primary‑care level, thereby moving toward more equitable access to healthcare.

---