On 29 September 2026, OpenAI announced Dots: always-on AI agents that work on your behalf around the clock instead of waiting for prompts. For Indian small businesses already using AI for captions and emails, this is a shift from assistant to coworker. If you are new to business AI, start with our AI tools for small business in 2026 and our view on agentic AI in marketing, then read on for what Dots changes.
Before giving any always-on agent access to business systems, read our AI agent governance guide on moving from pilot to production, and the AI agent security checklist.
What OpenAI Dots actually are
Dots are agents powered by GPT-6 Astra with their own cloud computer, their own browser, and connections to more than 4,000 apps through plugins. Unlike a chatbot that answers and forgets, a dot learns your preferences, keeps working toward your goals 24/7, and brings you finished work — sometimes before you ask. You reach it through ChatGPT on desktop, web, and mobile, in Slack and Teams, or on a voice call, with texting support coming.
The magic, in OpenAI own words from the launch, is work arriving done the way you would do it. Their examples include a bug reported in Slack that dots immediately investigate, a new design turned into a working app while the team handles feedback, and an early tester whose dot noticed a forgotten invoice, prepared it, and sent it after approval. For a small business, translate that to: quotations chased, reviews answered, stock alerts raised, and posts drafted — with you approving the consequential steps.
Personal dots versus specialist dots
OpenAI describes two flavours. Your personal dot works for you: it learns how you think, what good looks like, and your audience and standards. Specialist dots take on defined organisational responsibilities with their own identity, credentials, and access to systems of record — built with lessons from pilots across procurement, invoice processing, email marketing, customer support, and commercial contracting. OpenAI is also integrating specialist dots with Microsoft enterprise governance in Agent 365.
| Aspect | Personal dot | Specialist dot |
|---|---|---|
| Who it serves | You, learning your preferences | A team function with defined duties |
| Access | Apps you connect, your rules | Own identity and system credentials |
| Rollout | Pro and Business Premium plans | Focused enterprise pilots |
| SMB relevance today | High: follow-ups, drafts, monitoring | Later: once scaled beyond pilots |
For most Pune or Vadodara SMBs, the personal dot is the relevant starting point. Specialist dots matter once you have systems of record worth connecting — the same discipline as our first automation workflow advice: one clear process before fancy tooling.
You stay in control: approvals and safeguards
The launch puts unusual weight on control, which is exactly what business owners should demand. Dots start with built-in rules for when to act alone and when to ask. Custom Rules let you allow, require approval for, or block specific actions. An auto-review layer checks risky actions against your instructions before proceeding, and sensitive tasks such as changing a password always stay with you. You can watch background work in Activity View and redirect the dot as needed.
Two details matter for trust. When idle, a dot does proactive research using read-only tools — it cannot send messages, change app content, or control your browser or computer in that mode. And you choose which apps it can access through existing ChatGPT controls. Pair this with the consent habits in our AI content workflow and WhatsApp automation guide: the business decides what the agent may touch, in writing, before it touches anything.
Availability and cost: what is confirmed
Dots are rolling out to Pro and Business Premium users in eligible markets, with Enterprise beta where workspace admins enable it. Your first dot is included in Pro or Business Premium at no extra cost, with an allowance for deeper work and extended limits for the first month after launch. Conversations with your dot do not count toward ChatGPT usage limits, while tasks it starts in Codex or ChatGPT Work count as usual. Availability in India depends on eligible-market rollout, so check your plan rather than assuming access.
Treat this like any new channel: pilot one workflow, measure, then expand. Our AI prompts playbook gives your team reusable starting points while you wait for access, and AI customer support basics covers the narrow-first discipline that also applies to dots.
Honest limits SMBs should keep in mind
Dots are new, and new systems stumble. OpenAI itself says dots can still make mistakes and asks users to review consequential work. Early always-on agents will misread vague instructions, overstep unclear boundaries, and confidently complete the wrong task. The fix is the same discipline as any hire: narrow scope, written instructions, and checking the first dozen outputs before trusting the hundredth.
Cost is the second open question. The first dot is included, but deeper work runs on an allowance, and future dots and speed scaling will carry their own pricing. A small business should pilot within the included allowance and measure hours saved before spending more — the same ROI habit in our workflow automations guide. If a dot saves ten owner-hours a month on follow-ups and drafting, the maths is easy; if it merely drafts captions you rewrite fully, simpler tools suffice.
Data control deserves the same scrutiny. OpenAI states business and education workspace content is not used to train models by default, with personal-plan controls for the rest — read those settings before connecting shop data, and keep customer phone lists and bills out of any AI tool without a clear purpose, as our prompts playbook advises. A careful pilot now beats a rushed rollout later.
What Indian SMBs should do right now
You do not need Dots access today to benefit tomorrow. The businesses that gain most will be the ones with clean foundations.
- Write down repeatable workflows. List the weekly tasks a dot could own: enquiry follow-ups, review replies, quotation reminders, post drafting. If it is not written down, an agent cannot learn it.
- Clean app permissions. Review who can access your sheets, CRM, and social accounts today. A dot inherits the mess or the order you already have.
- Fix approval habits. Decide who approves quotes, refunds, and posts now, so Custom Rules mirror a real process instead of inventing one.
- Keep humans on exceptions. Complaints, pricing judgment, and promises stay with people — the same handoff rule in our chatbot guide.
Start with one workflow that stops leakage this month, exactly as you would with any automation pilot. Contact RadhaKrishna Web Solutions or call +91 70962 91214 to map the first process a dot — or simpler automation today — should own for your Pune, PCMC, or Vadodara business.
Frequently asked questions
What are OpenAI Dots in simple words?
Always-on AI agents powered by GPT-6 Astra with their own cloud computer that work toward your goals around the clock across ChatGPT, Slack, Teams, and voice, connecting to thousands of apps.
How much do Dots cost?
The first dot is included in Pro or Business Premium plans at no extra cost with an allowance for deeper work. Conversations do not count toward ChatGPT limits, while Codex or ChatGPT Work tasks it starts do count as usual.
Are Dots available in India?
Rollout covers Pro and Business Premium users in eligible markets, with Enterprise beta via workspace admins. Check your ChatGPT plan and OpenAI announcements for current India availability.
Is it safe to let a dot access business apps?
You control which apps it accesses, set Custom Rules for approvals, and monitor work in Activity View. Background research is read-only, and sensitive tasks always stay with you. Start with one low-risk workflow.
