On March 16, the China Academy of Information and Communications Technology (CAICT) officially launched its 2026 "Trusted AI-Agent" series evaluation. The initiative aims to foster high-quality development of AI agents and deepen their integration with the real economy, building a universally beneficial and trustworthy industrial ecosystem.
The evaluation assesses AI agents across two primary dimensions: fundamental technology and application capabilities. The fundamental technology assessment covers platform tools, technical capabilities, operational services, and engineering proficiency. The application capability assessment encompasses general tasks, industry-specific applications, and maturity evaluations. The first batch of evaluation results is scheduled to be released at a major industry conference in May 2026.