OpenAI is pushing AI beyond the single prompt with Dots, personal agents designed to stay active throughout the day. Instead of waiting for every next instruction, a Dots agent can continue working after the first request, turning a conversation into an ongoing process.
At the same time, Tencent has signed its largest overseas lease deal with Oracle to access about 100,000 advanced AI chips that were not available in China. Together, the two developments point to the same pressure shaping AI: smarter software needs more independence, while ambitious AI initiatives need access to powerful hardware.
Dots Keep Projects Moving After the First Request
Most AI interactions begin and end with a prompt. A user asks for something, receives a response, and then returns with another instruction. Dots changes that pattern by keeping track of an ongoing project instead of treating every conversation as a separate task.
A user could ask a Dots agent to follow a project, prepare material, or keep information updated as new details arrive. The agent can continue that work without waiting for the next prompt, giving the task a sense of momentum that ordinary back-and-forth conversations do not provide.
That shift matters because many projects do not fit into one request. Information changes, materials need preparation, and progress must continue after the user steps away. Dots are built for that flow, with each agent staying active throughout the day and continuing work after the initial instruction.
The result is an AI assistant that can remain connected to a task rather than simply answering questions about it. The user gives the direction, and the Dots agent keeps the project moving as new details arrive.
Cloud Computers Give Each Agent Room to Act
Each Dots agent runs through its own cloud computer. That setup allows the agent to use tools, browse the web, work with approved apps, and create files as it handles an ongoing assignment.
The cloud computer gives Dots a place to carry out work instead of limiting the agent to a reply inside one conversation. A task can involve gathering information, preparing material, updating files, or using approved applications, all within the same continuing process.
Dots can also work with products such as Codex and ChatGPT. The agents can communicate through services such as Slack and Teams, connecting their ongoing work with the tools people use to manage projects and share information.
- Dots can stay active throughout the day.
- Each agent runs through its own cloud computer.
- Agents can use tools, browse the web, and work with approved apps.
- Dots can create files and keep information updated as new details arrive.
- Agents can work with Codex, ChatGPT, Slack, and Teams.
These capabilities make Dots less like a chatbot waiting for questions and more like a continuing digital worker following a defined assignment. The important change is not only what the agent can do, but when it can do it: Dots can keep working without waiting for the next prompt.
Tencent’s Chip Deal Shows the Hardware Race
Software agents need computing power to run, and Tencent is securing that power through Oracle. The Chinese company has leased about 100,000 advanced AI chips from the US cloud provider in its largest overseas lease deal.
The chips were not available in China, making overseas access a central part of the arrangement. Tencent can use the hardware to support its AI initiatives, linking the company’s ambitions to the supply of advanced computing resources outside the country.
The deal adds a hardware dimension to the rise of autonomous AI agents. OpenAI’s Dots focus on giving software more room to act, while Tencent’s lease shows how access to chips can shape the ability to build and support AI systems.
That connection will keep drawing attention as AI work moves from isolated answers to ongoing processes. The more tasks an agent handles, the more important its cloud computer becomes. The more AI initiatives a company supports, the more urgent access to advanced AI chips becomes.
Dots now bring together continued attention, cloud-based tools, approved apps, file creation, and communication services in one agent model. Tencent’s deal shows the other side of that expansion: AI progress also depends on the hardware available to power it.
The next stage of AI will be measured by more than how well a system answers a prompt. It will also be judged by whether the system can follow a project, keep information current, and continue working after the user leaves the conversation. Dots are built around that future, while Tencent’s chip lease shows how much computing power that future demands.
Based on
