China Telecom's workplace artificial intelligence agent is now used by more than 1.5 million people, according to Media OutReach Newswire's account of an announcement made at the opening of PT Expo China 2026 on October 9. The product, called TeleAgent, is one of the largest enterprise AI agent deployments disclosed by a national telecom operator to date.
The TeleAgent AI Agent has also crossed the 200,000 daily-active-user mark, the company reported. That represents steep growth from just 27,000 registered users in mid-June, a span of under four months. TeleAgent was first unveiled earlier this year at the World Artificial Intelligence Conference (WAIC) 2026.
Liu Guiqing, Executive Director, President and Chief Operating Officer of China Telecommunications Corporation, announced the milestone at the trade expo. In his remarks, reported by Media OutReach Newswire, he framed the product as evidence that AI is shifting from understanding what people say to completing work on their behalf β a shift he described as moving from comprehension to execution.
The scale of the adoption matters because enterprise AI agents have struggled to move beyond pilots. According to China Telecom, the user figures point to autonomous systems capable of end-to-end task execution entering mainstream corporate use rather than remaining showcase demonstrations.
How TeleAgent is built and who uses it
TeleAgent is developed by China Telecom AI Technology Co., known as TeleAI, as a native AI application for the workplace. According to the company, the user base spans individual professionals, small and medium-sized enterprises, and large state-owned conglomerates.
The product comes in several editions. An enterprise SaaS version supports collaboration across multiple devices and customization for specific work scenarios. A separate private-deployment edition, released in mid-August, keeps models and data entirely inside customers' own local environments β a design aimed at organizations with strict data-sovereignty requirements.
TeleAgent runs natively on Windows, macOS, and leading Chinese enterprise operating systems including Kylin OS and UnionTech OS (UOS). According to the announcement, individual users need no complex server setup; they can drag and drop files into the workspace and let the agent carry out end-to-end work from there.
One of the most striking details in the announcement concerns the product's skill marketplace. That marketplace has grown from roughly 40,000 skills to more than 50,000, and 99 percent of those skills were built directly by end users solving their own productivity problems, reported Media OutReach Newswire. The figure suggests an unusually active user-generated ecosystem for an enterprise tool.
What early enterprise users report
China Telecom also shared results from an in-depth survey of key enterprise pilot and power users. According to the survey, 90 percent of respondents said the TeleAgent AI Agent has become integral to their daily routines. Document drafting, industry research, and automated data analysis account for more than half of all usage across the surveyed group.
On productivity, 79 percent of surveyed users cited significant gains, while 35 percent said the tool exceeded their expectations. The product scored a Net Promoter Score of 62.9 in the survey. Among the 74 percent of respondents who had previously used rival platforms, 89 percent rated TeleAgent on par with market leaders, while 16 percent rated it superior, according to the company.
The company offered a concrete workflow example. A consulting analyst described a recurring quarterly task β one that previously required exporting more than 20 reports from five separate systems, followed by half a day of manual data cleaning and charting β as now taking about five minutes. Users upload raw data, state their objectives, and receive finalized reports with interactive charts.
In another case, a regional sales and channel operations unit within China Telecom used the agent during a major commercial campaign to reduce daily cross-system consolidation work from hours to minutes, automating report distribution entirely, reported by the company.
Security, cost control, and language tuning
China Telecom framed TeleAgent's competitive edge around three pillars: security, efficiency, and results.
On security, the product is built with carrier-grade protection and implements a four-layer defense loop covering perception, reasoning, execution, and memory. A lightweight security sandbox isolates file access. According to the announcement, the product has achieved certifications from the China Academy of Information and Communications Technology (CAICT) and cleared 99.7 percent of 336 threat-evaluation scenarios run by QiAnXin and other leading security vendors.
On efficiency, TeleAgent runs on China Telecom's own Xingchen multimodal foundation model and its integrated cloud infrastructure. The company said it has pioneered "token-based operations" β an intelligent model router directs requests to the most cost-effective foundation models while context compression reduces latency and inference costs.
On results, the product is tuned specifically for Chinese-language workplace workflows, particularly complex spreadsheets, briefings, and presentations. It incorporates persistent long-term memory that adapts to user habits over time. According to the announcement, in IDC's technical evaluation of Chinese enterprise-grade general-purpose agent products, TeleAgent scored 6.85 and ranked in the top three nationwide, with cost efficiency, security, and task performance scoring well above industry benchmarks.
An expanding agent ecosystem
TeleAgent serves as the enterprise flagship for China Telecom's unified group-wide agent architecture. At its core is an agent hub called "Xing Xiaochen," which drives multimodal intent recognition, task planning, and tool orchestration under a "build once, reuse across the network" framework, according to the company.
The ecosystem around the product is widening. China Telecom has integrated capabilities across navigation, corporate travel, lifestyle, and weather services, and 60 agent applications from 44 external partners are currently undergoing technical integration, reported by Media OutReach Newswire. The company aims to extend these agentic capabilities across broader enterprise ecosystems, working through China Telecom Global to support multinational organizations.
Professionals across marketing, engineering, legal, and consulting are delegating routine paperwork, ledger consolidation, and contract risk reviews to the agent, according to the announcement.
The growth trajectory puts the TeleAgent AI Agent among the fastest-scaling enterprise AI agents in the market. For comparison, other recent milestones in the agentic AI space include rapid enterprise adoption of agents embedded in everyday work tools. See this reporter's earlier coverage of Google's Gemini agent gaining its own Gmail and Drive access, and of Anthropic cutting internet access for its internal AI agent evaluations after safety incidents.
More detail is available in the original October 9 announcement on Media OutReach Newswire, with additional coverage from LiveNews.co.nz and SUPERADRIANME.
Comments 0
No comments yet. Be the first to share your thoughts!
Leave a comment
Share your thoughts. Your email will not be published.