OpenAI introduces Dots, an enterprise-focused agent platform

OpenAI has introduced Dots, a new agent platform designed to function as a digital coworker. Unlike consumer-oriented AI agents aimed at personal shopping or general tasks, Dots is built primarily as business software that can interact with other desktop applications and cloud-based systems.

Platform Architecture and Capabilities

Dots features a virtual machine interface where users can observe the agent interacting with software. The platform provides out-of-the-box access to applications such as Blender and GIMP, and can integrate with a user's local desktop environment through the ChatGPT application. Users can also communicate with their Dots via voice calls to coordinate multi-step projects.

OpenAI has made the platform available initially to users on its highest-tier accounts, including the $100-per-month Pro tier, signalling an enterprise-first approach to the rollout.

Practical Testing and Limitations

Testing the agent across various personal and professional tasks yields mixed results. Due to its cloud-based infrastructure and browser footprint, Dots occasionally encounters difficulties with automated security checks, such as continuous mouse-hold verifications or login loops on third-party sites, requiring manual intervention from the user.

When applied to structured professional workflows—such as website redesigns, managing video editing assets, and deploying backend changes—the agent demonstrates higher reliability. Users can delegate multi-step tasks while monitoring progress remotely through mobile notifications.

Outlook for Enterprise AI

Positioned as a productivity tool rather than a personalized consumer assistant, Dots represents a shift toward software-driven automation in the workplace. While security bottlenecks and pricing structures may limit initial adoption for casual users, the platform offers a blueprint for how AI agents might handle complex, multi-application workflows in professional environments.