In partnership with
Did you know that we have LinkedIn, Instagram and X accounts
that you can follow?
Hi {{first_name|everyone}}, OpenAI's new "dots" are AI agents that keep working while you're off doing something else, and the interesting part is how much they're allowed to do on their own. After that, we've got a take on what makes agents actually useful, a few tools worth a look, and something fun someone built with AI.
Let’s get right into it!
🤿 DEEP DIVE
AI Coworker That Doesn’t Just Wait Around
Most AI tools still work like this. You ask for something, they respond, then they wait.
Dots are built around a different idea.
OpenAI describes them as always-on agents that can keep working toward your goals, use their own cloud computer, connect to thousands of apps through plugins, and come back when they have work ready or need a decision from you.
And the more you work with one, the idea is that you have to explain yourself less.
You can hand over the smaller stuff
You can give a dot several projects without managing each one as a separate conversation.
OpenAI’s examples make the distinction pretty clear.
A developer’s dot can watch customer feedback, identify smaller fixes, build and test them, then bring back finished pull requests with videos showing the changes.
A scientist’s dot can rerun analyses when new data arrives and update figures and explanations.
A sales lead’s dot can keep checking changing customer requirements, update a proposal, and flag unresolved concerns.
The point isn’t really that dots can perform one impressive task. It’s that the work can keep moving while you’re doing something else.
But it doesn’t get free rein
When you aren’t actively working with your dot, it can look through connected apps for ways to help. OpenAI calls this proactive research.
But that background access is restricted to read-only tools. It can inspect information, but it can’t quietly send messages, edit app content, or take control of your browser or computer.
For actions that do affect accounts or share information, dots use an auto-review system to decide whether the action can proceed, needs your approval, or has to be completed by you.
You can also create Custom Rules that allow, block, or require approval for particular actions.
So this is less about letting an AI loose on everything and more about giving it room to keep working inside boundaries you can inspect and change.
OpenAI is also previewing specialist dots for companies. Unlike your personal dot, these would have their own organizational identity, credentials, hardware, and access to company systems so they can take on defined responsibilities.
The quick take
Dots are always-on AI agents powered by GPT-6 Astra with their own cloud computers.
They can work across multiple projects and connect to more than 4,000 apps through plugins.
Background proactive research is limited to read-only access.
Actions that affect accounts or share information can require review or approval.
OpenAI is also testing specialist dots designed for defined roles inside organizations.
🤝 FROM OUR FRIENDS
A quick share we thought you might find useful
A fast answer and a coordinated one are not the same thing. When AI resolves a B2B customer issue without looping in the teams who have to deliver on it, you get confident responses nobody actually signed off on.
A new briefing paper from Harvard Business Review Analytic Services, sponsored by Front, examines the coordination gaps that open up when transactional AI tools meet multi-team B2B service, and how leading companies are using AI to close those gaps instead of widening them.
Read the briefing paper for the questions to ask before your next AI investment.
🔥 WHAT’S THE BUZZ
What people are building, sharing, and talking about
I keep coming back to this idea that the useful AI layer might not be the agent itself, but all the stuff around it that lets it actually do a job. The tools, rules, context, permissions, basically the boring bits that make it reliable in one specific domain.
⚒ TOOL SNAPSHOTS
Futuristic tools within AI, no-code, and productivity
Give your AI agents some memory of what you actually did.
Why it’s useful: Handy when the page, conversation, or meeting detail you need is somewhere in your recent screen history, without connecting every app individually.
Check your website numbers yourself, or just ask your agent.
Why it’s useful: A nice fit if you want straightforward analytics that your AI tools can also query, without maintaining a separate analytics workflow for them.
📈 ZenABM
Run more of your LinkedIn ads from the AI tool you already use.
Why it’s useful: Useful for teams that would rather create campaigns, inspect performance, and handle reporting without constantly jumping back into Campaign Manager.
🤖 AI MADE THIS
Interesting and inspiring creations made with AI
An interactive history of Earth built in about 30 minutes
A creator used GPT-6 Astra to build an interactive history of Earth in roughly 30 minutes. The result is a 3D browser experience with moving continents, changing oceans, ice and terrain, showing AI being used to build an explorable educational interface rather than just generate text or images.
AI used
GPT-6 Astra, used through the command line. The site was hosted on Vercel.
ℹ️ ABOUT US
The Intelligent Worker helps you to be more productive at work with AI, automation, no-code, and other technologies.
We like real, practical, and tangible use-cases and hate hand-wavy, theoretical, and abstract concepts that don’t drive real-world outcomes.
Our mission is to empower individuals, boost their productivity, and future-proof their careers.
We read all your comments - please provide your feedback!
Did you like today's email?
What more do you want to see in this newsletter?
