← Back to Insights

OpenAI Dots: Always-On Agents for Pro and Business Teams

At DevDay on Sep 29, 2026, OpenAI launched Dots—always-on agents powered by GPT-6 Astra that work on their own cloud computer across connected apps, ChatGPT, Slack, and Teams. For owners on Pro / Business Premium / Enterprise, the win is background progress with approval gates—not unattended autopilot.

OpenAIDotsGPT-6 AstraAI agentsChatGPT BusinessoperationsAuto-reviewCerpaMedia Insights

OpenAI Dots: Always-On Agents for Pro and Business Teams

What happened

On September 29, 2026, at OpenAI’s DevDay, the company launched Dots: always-on personal agents powered by GPT-6 Astra. OpenAI describes them as agents that learn how you work, use their own cloud computer and connected tools (OpenAI and press cite browser access and thousands of supported apps), and keep making progress between chats—bringing work back for review. You can message or call a Dot in ChatGPT (web, mobile, desktop); Dots also work in Slack and Microsoft Teams with carried-over context; texting support is listed as coming next. Access is rolling out to ChatGPT Pro, Business Premium, and Enterprise in eligible markets (Enterprise when enabled by a workspace admin). OpenAI says you manage app permissions and Custom Rules; actions that affect accounts or share information go through Auto-review, and sensitive steps (e.g., changing a password) stay with the user. Press coverage (TechCrunch, The Verge, Axios) places Dots as OpenAI’s answer to Meta’s Muse-style always-on assistants, with an initial one-Dot-per-user rollout and specialist Dots for org roles under test.

Who it helps

  • Owners and managers already on ChatGPT Pro or Business Premium who want background follow-through on decks, FAQs, research, and drafts—not another blank chat box.
  • Sales, ops, and content leads who work across Slack/Teams and need an agent that shows up where the team already talks.
  • Tech-services firms advising clients on agent adoption who must pair capability with approval policy from day one.

How to apply

  1. Confirm plan eligibility (Pro / Business Premium / Enterprise) and create one Dot with a single owned responsibility (e.g., “refresh investor deck figures” or “draft social from approved transcript”).
  2. Connect only needed plugins; set Custom Rules that block unsolicited sends and require approval for email, purchases, and deletes.
  3. Use Activity View (desktop) to review ongoing tasks daily for the first two weeks.
  4. Keep proactive research in read-only mode for discovery; require a human gate before any external action.
  5. Measure outcomes as “hours of prep saved + error rate,” then decide whether to expand scope—or wait for multi-Dot / specialist Dot availability.

How CerpaMedia can help

CerpaMedia treats Dots as an operations tool, not a magic employee. A Technology Strategy Call compares Dots vs. alternatives against your current stack and risk tolerance. Build configures permissions, Custom Rules, and a pilot workflow. Automation formalizes draft → review → approve loops so agents move work forward without sending or spending on their own.

Sources

  • OpenAI / ChatGPT, Dots product page: https://chatgpt.com/features/dots/
  • OpenAI Help Center, Dots privacy, security, and safety FAQs: https://help.openai.com/en/articles/20001529-dots-privacy-security-and-safety-faqs
  • TechCrunch, OpenAI launches Dots, its bubbly agentic avatar (Sep 29, 2026): https://techcrunch.com/2026/09/29/openai-launches-dots-its-bubbly-agentic-avatar/
  • The Verge, OpenAI launches Dots, its Muse competitor (Sep 29, 2026): https://www.theverge.com/ai-artificial-intelligence/1002033/openai-dots-launch-muse-competitor
  • Axios, OpenAI launches dots AI agents, seeks to address safety concerns (Sep 30, 2026): https://www.axios.com/2026/09/30/openai-dots-ai-agent-safety