What are OpenAI's "Dots"?

OpenAI's "Dots" are new always-on artificial intelligence agents launched at DevDay 2026. They are designed to perform multi-step tasks for users, connect with existing applications, and learn user preferences over time.

How do "Dots" differ from traditional chatbots?

Unlike single-turn chatbots, "Dots" feature an "always-on" design, continuously monitoring the web and working towards assigned tasks. They also pull context from connected apps and proactively assist users.

What is the purpose of "specialist Dots" for enterprise customers?

"Specialist Dots" are tailored for business clients to handle specific professional tasks like accounting, email marketing, and legal analysis. This aims to integrate AI agent "coworkers" into enterprise workflows.

What security and privacy features do "Dots" include?

"Dots" are designed to seek explicit user approval for sensitive actions like installing software or changing passwords. Users can also set "Custom Rules" to define boundaries and tasks requiring direct permission.

OpenAI Introduces 'Dots' AI Agents at DevDay 2026, Rivaling Meta's Muse

Q: Who can access "Dots" initially?

The rollout of "Dots" began for subscribers to ChatGPT's Pro tier, which costs $100 per month. OpenAI typically introduces new features to its higher subscription tiers first.

OpenAI has launched "Dots," a new line of always-on artificial intelligence agents, at its DevDay 2026 event in San Francisco on Tuesday, according to Wired. The introduction of Dots marks OpenAI's entry into a competitive market for personalized AI tools, challenging Meta's Muse agent, Business Standard reported.

The Dots agents are designed to perform tasks on behalf of a user with minimal supervision and are depicted as customizable "cute blobs," Wired reported. They are built to connect with existing applications and manage multi-step projects.

Launch Details and Features

The new AI agents are powered by OpenAI's GPT-6 Astra model, Wired stated. A key characteristic of Dots is their "always-on" design, which differentiates them from traditional single-turn chatbot interactions. These agents are engineered to continuously monitor the web and work towards completing assigned tasks. They also gather context from linked applications and progressively learn user preferences over time.

During a demonstration, OpenAI showcased a Dot assisting a user with dinner plans. The agent proactively identified from the user's calendar that they would be working through dinner and subsequently offered two GrubHub food delivery options, including their respective prices. The user then selected their preferred meal and specified the desired ordering time. In another example, a Dot collaborated with a user to facilitate the launch of a new website, Wired reported.

Users can interact with their Dots through various platforms, including ChatGPT, Slack, and Microsoft Teams, with contextual information shared across these modes. Additionally, subscribers to the ChatGPT Pro tier can join a waitlist to communicate with their Dots via iMessage or RCS messaging on Android devices, according to Wired.

The rollout of Dots commenced today for users subscribed to ChatGPT's Pro tier, which costs $100 per month. OpenAI typically introduces new features to its higher subscription tiers first before making them available to a wider user base. Initially, users will be able to control one Dot, though the company anticipates eventually enabling control over multiple agents simultaneously, Wired reported.

Enterprise Focus and Competitive Landscape

The launch of Dots is part of OpenAI's broader strategy to enhance its offerings for enterprise clients, Business Standard reported. At DevDay, OpenAI CEO Sam Altman announced "specialist Dots" specifically tailored for business customers. These specialized agents are designed for particular tasks such as accounting, email marketing, and legal analysis, further integrating AI agent "coworkers" into professional workflows, Wired added.

The introduction of Dots positions OpenAI directly against Meta, which recently saw significant user adoption for its Muse agent. Wired noted that Meta's Muse agent reached a large user base and quickly ascended to the top of smartphone app download charts, indicating a "breakthrough moment" for personal agents. The market has also seen early adoption of tools like OpenClaw by users in Silicon Valley earlier this year.

Security and Privacy Considerations

OpenAI has incorporated safety mechanisms into the Dots agents. For sensitive actions, such as installing software or changing a password, Dots are designed to seek explicit user approval before proceeding. Users also have access to a "Custom Rules" tool, which allows them to define specific boundaries that agents should not cross and tasks that require more direct permission, Wired reported.

However, as users connect their personal data and integrate these agents into their lives, potential security and privacy risks remain a significant consideration. Wired advised caution, noting that while OpenAI has built safety guardrails, users should be mindful that these tools could inadvertently share unintended information or be exploited by online adversaries to perform unwanted actions. The publication also highlighted that if a user's OpenAI account is configured to permit model training on their data, this setting will also apply to interactions with the new agents.

The company's move into always-on AI agents aims to capitalize on increasing demand for tools that can manage daily tasks with minimal human oversight, Business Standard reported.

Future Outlook

While the initial rollout allows for control of a single Dot, OpenAI plans to expand this capability to allow users to manage multiple agents concurrently in the future, according to Wired. The company's strategy suggests a continued focus on developing sophisticated AI agents that can adapt and learn from user interactions, potentially transforming how individuals and businesses manage their digital workflows.