Dots are single, private agents that run on their own cloud computer, powered by GPT‑6 Astra, can connect to thousands of apps, and are available only to paying ChatGPT customers at launch.
Dots are persistent, chat-based agents powered by GPT‑6 Astra that act on your behalf across apps and devices.
At launch, Dots can connect to over 4,000 apps, run on their own cloud computer, and are limited to one per paid ChatGPT user.
OpenAI built multiple safety layers: activity runs on a separate cloud instance, sign-ins pause model activity, Auto-review checks actions against rules, and users can set custom restrictions.
# What Dots are and how they work
# Capabilities and use cases Dots can connect to more than 4,000 apps at launch, including ChatGPT, Slack, and Teams. They can also connect to your devices if you grant permission. OpenAI says dots can perform anticipatory actions based on past interactions — for example, preparing an invoice you forgot to file and sending it after your approval, or monitoring customer feedback and fixing small code issues for an app you're developing.
OpenAI claims dots can run many concurrent tasks without slowing down because each dot runs on its own cloud instance. You can view that cloud computer at any time to see what the dot is doing and review its activity.
# Interaction model Dots are chat-first. You send messages in ChatGPT, Slack, or Teams and the dot replies with updates, questions, or proactive messages when something relevant happens. OpenAI calls this proactive research: the dot works in the background and reaches out when it detects developments that matter.
# Safety, privacy, and controls OpenAI describes several safeguards:
OpenAI also states that dots include protections against prompt injection and other malicious instructions, but the company acknowledges dots can still make mistakes.
# Availability and limits
# How Dots compare to other agents The article places Dots in the context of other consumer and workplace agents released this year. It notes recent entries such as Meta's Muse and Apple's Siri AI. Compared with those, OpenAI's approach is individualized: each user gets a single, private dot running on its own cloud instance and tied to the user's permissions and rules.
# What to expect next OpenAI intends to add the ability to create multiple dots later and continues to develop safety and privacy measures. For now, Dots aim to automate repetitive or time-sensitive work while leaving final approvals and sensitive sign-ins under user control.
# Bottom line Dots are designed as a private, always-on assistant that can take anticipatory actions across thousands of apps, with layered safeguards and user controls. At launch they are limited to paying ChatGPT users and to a single dot per account, and OpenAI warns they are not error-proof.
At OpenAI's annual developer conference, the company made 20 different announcements, but none bigger than the new AI agent, Dots.
OpenAI just announced a new suite of AI agents called “dots,” and they look like something you might find in an animated kids show. Dots were the star among more than 20 announcements made at OpenAI’s fourth annual developer conference, DevDay , which took place on September 29 in San Francisco. According to an officia
AI giant launches 'dots' to rival Meta's Muse and Google's Gemini Spark.
Meta’s Muse agent was born to shop , while Gemini’s Spark can help you book airline tickets or plan your weekly meals. But while Muse and Spark act like personal assistants, OpenAI’s new AI agents are designed to be, well, another “you” in the office. Dots, which OpenAI boss Sam Altman just unveiled at the company’s an
AI (Artificial Intelligence) နည်းပညာက အá€á€¯á€¡á€á€»á€á€”်မှာ နေရာá€á€á€¯á€„်းမှာ ရှá€á€”ေပါပြီዠဒါပေမဲ့ “AI ကá€á€¯ ဘယ်ကနေ စလေ့လာရမလဲአအမြန်ဆုံး á€á€á€ºá€™á€¼á€±á€¬á€€á€ºá€¡á€±á€¬á€„်â
Loading more related stories...
Open the app view to save this story, compare related coverage, and continue from the same source.