AgiBot's Lingxi X2 robot processed 46,686 patient inquiries at a Beijing hospital
Read more
Pandaily
pandaily.com

AgiBot's Lingxi X2 robot processed 46,686 patient inquiries at a Beijing hospital

The Lingxi X2 robot from AgiBot has become a permanent staff member at the 안내 desk of Peking University Hospital in Beijing. This project was implemented in collaboration with Ant Group's health application, Afu.

At this desk, Lingxi X2 utilizes a medical assistant based on the Afu application, which runs on Ant Group's large medical model. The robot answers standard questions using a humanoid voice and gestures, taking over frequently repeated inquiries that typically burden staff working at the busy hospital lobby desks.

According to company data, as of September 30th, the robot operated for 66 working days at the outpatient navigation desk, conducting 46,686 interactions. This equates to approximately 707 interactions per day, with a peak of 1,420 interactions on a single day.

The robot's functionality includes directing patients to the correct department, explaining visit procedures, answering questions about health insurance policies, and providing indoor directions within the hospital. The system is also integrated with the hospital's private knowledge base, allowing it to provide standardized answers based on the institution's internal information rather than general data.

It is important to note that the robot's role is navigation, not medical treatment. It supports desk staff with basic questions but does not engage in patient diagnosis or treatment. The companies describe this implementation as a joint research into intelligent medical services, planning to use the acquired interaction and operational experience in a hospital setting to create a scalable model.

Lingxi X2 is a semi-sized humanoid from AgiBot, designed for high-interaction environments where a compact and accessible appearance is more important than heavy manipulation. The hospital lobby fits this profile: it features high visitor traffic, many repetitive questions, and requires quick, consistent responses.

The deployment of the robot in the hospital complements AgiBot's efforts to enter public service roles, following previous programs in campuses and theme parks. For Afu, this means transitioning Ant Group's medical assistant from a phone screen to a physical presence at the patient's first point of contact.

Similar stories

Purple Mountain Laboratories and China Mobile Unveil Platform for Connecting AXP Agents
Read more
pandaily.com

Purple Mountain Laboratories and China Mobile Unveil Platform for Connecting AXP Agents

Purple Mountain Laboratories has launched the AXP platform, designed to facilitate communication between artificial intelligence (AI) agents, which operates based on deterministic networks. The presentation of this platform took place on October 16th at the Tenth Conference on Future Network Development in Nanjing. China Mobile, Jiangsu Future Network Group, and Jiangsu Future Networks Innovation Institute were involved in its development.

The developers position AXP as the industry's first agent connection platform built on deterministic network principles, although this claim has not yet been independently verified. A deterministic network implies the network's ability to guarantee bounded and predictable traffic delivery within a specific area, unlike the best-effort behavior of standard internet connections.

AXP applies this concept to a growing number of software agents that need to locate and invoke each other across various organizations and regions. According to an official announcement, the platform is capable of registering up to one million agents and simultaneously supporting 100,000 active agents.

The platform performs joint planning of computational and network resources, handles transformations between different protocols used by heterogeneous agents, detects failures, and switches within seconds. Furthermore, it enables semantic addressing, allowing an agent to be located by its functionality in approximately one hundred milliseconds.

Operating on CENI, China's national large-scale experimental site for future networks, the team reported that AXP successfully implemented high-precision agent addressing across different domains over distances exceeding 4,000 kilometers and reduced end-to-end task execution time by more than 20%, although the baseline for this comparison was not specified.

The presented materials did not specify which agent protocols AXP transforms. Information regarding support for widely used open standards such as Model Context Protocol (MCP) or Agent2Agent (A2A) was not disclosed, nor was an explanation provided for the acronym AXP.

Purple Mountain Laboratories stated that it has begun joint testing and industrial verification with telecommunications operators, communication equipment manufacturers, and a major steel enterprise across multiple scenarios. The date of commercial availability or pricing policy has not been announced.

Popular