OpenAI has updated ChatGPT's plug-ins by adding dedicated sidebars, interactive panels, file viewers, and support for automations. Basically, ChatGPT stops being just a chat and starts looking like a real application, with fixed spaces and flows that run even when you're not typing commands.
For those managing business processes, the key takeaway is this: some tasks that currently require a separate tool — like management software, a notification aggregator, or a trigger system — could be handled directly within ChatGPT, without additional integrations. This isn't a generic promise: it's the direction OpenAI is heading with this release.
The article analyzes what the new features really do, who should consider them right away, and where it's better to wait before touching existing workflows.
From chatbot to work environment: what has changed in plug-ins
Until yesterday, ChatGPT plug-ins were extensions that responded to a prompt and then disappeared. With the update announced by TechCrunch on September 29, 2026 , OpenAI changes the model: each plug-in can now have a dedicated sidebar — meaning a fixed space in the sidebar of the interface — and interactive panels that remain visible while you work.
There are also file viewer , integrated document viewers that allow you to open and consult files directly in the plugin panel, and a system of improved discovery , meaning simpler navigation to find available plug-ins without having to remember them.
The most relevant novelty for those managing processes is the automation support : flows that activate autonomously, without the user having to send a message. ChatGPT can perform an action at a fixed time, in response to an external event, or upon completion of another step in the flow.
Where this changes something in an SME — and where it doesn't
The useful question is not 'is it interesting?' but 'does it replace something I already pay for or manage with difficulty?'. Some areas where the answer might be yes:
- Receiving and classifying documents : a plug-in with a file viewer and automation can read an incoming PDF and catalog it without manual intervention.
- Recurring notifications and reports : instead of setting up a scheduled export on a separate tool, the workflow lives inside ChatGPT.
- Simplified internal interfaces : an interactive panel can become the access point for data or actions for those in the company who don't use technical tools.
Where it's worth waiting instead: if your critical processes are already running on a CRM or ERP with stable integrations, there's no point in dismantling them to experiment. ChatGPT automations are new, and third-party plug-ins need to update to take advantage of them. The ecosystem will stabilize in the coming months.
The comparison with AI agents you already know
This update is part of a larger movement. On the Microsoft front, Copilot Autopilot brings cloud agents with variable pricing and a different consumption model than OpenAI's. The practical difference: Copilot is integrated into the Microsoft 365 ecosystem, so it's more immediate for those who already use Teams and SharePoint; ChatGPT plug-ins are more open but require software providers to build the integration.
On the voice and mobile front, OpenAI had already taken a similar step: ChatGPT mobile has introduced voice-powered agentic tasks for Pro and Plus plans . Plug-in automations are the desktop and more structured version of the same concept.
It's also worth remembering that AI agents amplify what they find: if the data they work with is wrong, the result is wrong more efficiently. Incorrect audience data and AI agents are a combination that can cause serious damage — a principle that also applies to automations in plug-ins.
Requirements and concrete steps to evaluate it this week
No development skills are needed to use the new plug-ins. However, some conditions are required:
- A ChatGPT plan that supports plug-ins (check the settings section of your account).
- Plug-ins updated by their respective providers: not all have implemented sidebars and automations yet.
- A mapped business process: know exactly which activity you want to automate, with defined inputs and outputs, before touching any settings.
The most useful operational step this week is to open the ChatGPT plug-in section, check which tools you already use have released compatible updates, and choose just one low-risk process to test the sidebar or a simple automation on.
If you are evaluating how AI integrates into your workflows more broadly, the starting point is gathering insights on automation and AI agents that we keep updated on these topics.
The typical mistake: treating the plug-in as a finished app
Those who adopt these features right away tend to make a specific mistake: they build an elaborate workflow, put it into production, and then discover that a plugin or ChatGPT update has changed something in its behavior. Automations in AI environments aren't as stable as traditional software — they change with the underlying models.
The rule of thumb: keep critical flows on tools with declared SLAs and use ChatGPT for tasks where a temporary interruption won't block operations. Experiment, but within defined boundaries.
A useful comparison is to look at how other players are building similar integrations: Shopify has opened checkout to AI agents via browser with an approach that keeps responsibility layers separate — a model worth studying also to understand how to structure your own automations in ChatGPT.
Related articles
Discover more articles exploring similar topics, selected to offer you a more complete and stimulating perspective. Each piece of content is carefully chosen to enrich your experience.
