China Telecom recently launched a promotion through TeleAgent offering a free cloud computer with a licensed Windows environment and 4 cores and 8GB of memory. In actual testing, WorkBuddy can independently search the web, generate drafts, and save files in the cloud, so tasks no longer stop when a local computer is shut down. This matters because competition among AI agents is shifting from producing good answers to executing tasks reliably and remaining online for the long term.
What this is
The bundle has two parts: a free cloud computer provides a continuously available Windows environment, while an AI agent such as WorkBuddy—software that can autonomously invoke tools and complete multistep tasks—handles the actual work. After claiming the machine, users install WorkBuddy and sign in to the same account on both their phone and the cloud computer. They can then assign tasks such as drafting articles and processing spreadsheets from their phone, with files saved in the cloud.
In practice, WorkBuddy does not generate a full draft immediately. It first searches online, submits an outline and options for confirmation, and continues writing only after the user makes a selection. This reflects a collaborative model in which people approve key decision points while the agent executes the work. The free resource, however, comes with conditions: users must sign in at least once within seven days of claiming it, or their account may be reclaimed.
Industry view
We believe that putting cloud computers, AI agents, and loyalty points behind a single operator entry point lowers the barrier for ordinary users trying persistent background work. For businesses, this lightweight option can handle low-risk tasks such as preliminary content drafts and research organization; for individuals, it provides a background workspace that does not tie up local devices.
The risks are equally clear. The seven-day login requirement means the resource cannot be treated as infrastructure for long-term daily work. Cloud files raise questions about account permissions and data security, while generated results may still contain factual errors. If tasks involve customer records, financial data, or unpublished commercial information, businesses must first verify data retention, access controls, and exit procedures rather than focusing only on the word “free.”
Impact on regular people
For enterprise IT: Cloud desktops, software deployment, and account configuration once required separate procurement and setup. Now an operator can package them together. Pilots are easier, but before formal deployment, organizations still need to calculate long-term costs and assess permission and operational requirements.
For individual professionals: Tasks such as drafting articles and organizing research can run in the cloud while local devices remain available for other work. Its value lies in reducing idle time and device switching, not in fully replacing jobs or professional review.
For the consumer market: A free cloud computer lowers the price barrier to trying AI agents and may encourage operators to bundle cloud resources, loyalty points, and agent services. Users should distinguish temporary promotions from sustainable product capabilities.