China's StarWhisper telescope earns place in Stanford AI Index

China's StarWhisper Telescope has been cited in Stanford University's AI Index Report 2026 as a leading 2025 case study for AI agents in physics, astronomy, chemistry and materials science.

Led by the National Astronomical Observatories of the Chinese Academy of Sciences (NAOC), StarWhisper explores how large models and AI agents can be integrated with telescope control systems, helping with scientific observations under varying environmental and equipment conditions.

To date, the agents have flagged eight very early supernova candidates, two of which triggered follow-up observations when weather permitted.

AI is moving beyond data analysis into the operation of scientific instruments and research. The NAOC launched its AI for Astronomy program to seize this shift, and it's drawing international recognition for China's AI-driven observation work.

The program spans data processing, scientific discovery, telescope control and coordinated observations across facilities. The aim is to replace scattered algorithms with one system that covers the research process.

The team reports progress in three areas: digital simulation of research-grade telescopes, coordination between scientific models and observation-control systems, and autonomous observation by AI agents.

Research-grade telescopes are complex, and their observations depend on the optical tube, mount, camera, sensors, controls and weather. For AI agents to play a real role, they need an environment that can sense equipment status, understand conditions and test control strategies.

Observing time is expensive and hard to come by. Instead of using the telescope itself, the NAOC team built a virtual version that simulates its components, sensor states, environmental changes and workflows – a testing ground for control strategies, agent training and observation procedures.

They're also connecting scientific models, environmental models, telescope controls and AI agents, so the agents can use the tools and devices on hand.

They can weigh target priorities, the telescope's real-time status and visibility windows, then decide what to observe, plan it and check the result.

(With input from Xinhua)