Introduction: the chore I stopped doing by hand
I manage WordPress sites every day, and a surprising amount of that work is not design or strategy. It is chores: draft a post, fix a typo on a page, swap a photo, check which posts are gathering dust, update a plugin description. None of it is hard. All of it eats the afternoon in small bites, because every job means opening the dashboard, finding the right screen, and clicking through the same menus.
So when WordPress.com opened its AI agent connection to free-plan sites through November 2026, and Jetpack announced the same free window for self-hosted sites until 30 November 2026, I paid attention. This is not another AI writing plugin that lives inside your dashboard. It is the reverse: your ChatGPT or Claude chat, connected to your site, able to read it and change it while you type in plain language.
What is actually free, and until when
Let me get the dates straight first, because this is where most of the confusion lives.
On WordPress.com, connecting an external AI agent — ChatGPT, Claude, or any agent that supports MCP — is included with paid plans, and supported on the free plan through November 2026. Jetpack says the same thing for self-hosted sites connected to WordPress.com: until 30 November 2026, connecting ChatGPT or Claude to your Jetpack site is free, and you do not need a paid Jetpack plan.
Two things are easy to misunderstand here. First, your ChatGPT or Claude subscription is separate. The free window covers the WordPress side of the connection, not your AI app's plan — you keep using whatever ChatGPT or Claude plan you already have. Second, after 30 November, WordPress and Jetpack say their pricing pages will show which plans include AI agents. Everything the AI made for you stays on your site either way. Nothing gets deleted when the window closes; you just may need a paid plan to keep the connection live.
How the connection works, in plain language
The technology underneath is MCP, the Model Context Protocol — an open standard that lets an AI app talk to other software in a structured way. Think of it less like a chatbot that writes text and more like a remote control with rules. When you connect, the AI can see your site's posts, pages, media, and settings through declared tools, and it can only do what the connected WordPress user account is allowed to do.
For a WordPress.com site, the official route is the WordPress.com plugin for ChatGPT or the WordPress.com connector for Claude. For a self-hosted site, the route is Jetpack: install Jetpack free, connect it to WordPress.com, and then add the same WordPress.com app in ChatGPT or Claude. Coding agents work too — Claude Code, Cursor, VS Code, and Codex — with a one-line setup command that Jetpack publishes, and any app that supports MCP can connect through the same settings.
Connecting it: the steps I actually took
First, I made sure the site was running Jetpack and connected to WordPress.com. If you have never installed Jetpack, it is free from the plugin directory, and the connection step walks you through signing in with a WordPress.com account. That account is the bridge: it is the account you will sign in with again inside ChatGPT or Claude.
Second, in ChatGPT I added the WordPress.com app, and in Claude I added the WordPress.com connector. Both ask you to sign in with the WordPress.com account linked to your site and approve the connection. At this step you choose the level of access — read only, or full access. I started with read only, and I recommend you do the same. You can widen access later, once you have seen what the agent does with read access and you trust it with more.
Third, I asked it something small and verifiable: list my five most recent posts and tell me which ones have no featured image. That is a deliberately boring first task. It tests whether the connection actually sees your real site, and the answer is something you can check in thirty seconds in your dashboard. It passed. Only then did I move on to edits.
What I asked it to do, and what actually worked
Once the connection was live, I treated it like a junior assistant: good at defined tasks, bad at judgement calls, and always worth checking.
Drafting and editing posts worked best. I asked it to draft a post from a short brief, give it a category and tags, and save it as a draft — not publish it. It did that cleanly, and the draft was waiting in the dashboard for a human read-through. I asked it to fix a typo and tighten a paragraph on an existing page. Also clean. The useful pattern here is the one I keep coming back to: let the agent do the first pass, and keep publish in human hands.
Where it struggles, honestly
Two limits showed up quickly, and neither of them is a deal-breaker — but they decide whether this feels magical or messy.
First, the agent is only as good as your instruction. "Improve my site" produces vague, confident changes you then have to undo. "Draft a 400-word post about our new service, in a plain tone, save it as a draft, do not publish" produces something usable. The more specific the brief, the better the result. This is not a flaw in the tool so much as a fact about working with any assistant, human or AI.
Second, it does not know your taste, your brand voice, or what you would never publish. It will happily suggest a title that sounds like every other AI-written blog on the internet. That is why I keep drafts as drafts. The agent saves time on the first 80 percent; the last 20 percent — the voice, the judgement, the decision to publish — is still yours, and it should stay yours.
The safety checklist before you grant full access
I am going to be blunt here, because this is the part people skip. Connecting an AI to your site is connecting a powerful tool to your site. Do it properly and it is fine. Do it carelessly and you will regret it.
Start with read-only access and prove the connection works. Then, before you switch to full access, make a backup you can actually restore — not a backup plugin you installed and never tested, but a restore point you trust. If your host offers one-click backups, take one now.
Use a dedicated user account for the connection where you can, rather than your main administrator login. An editor-level account can draft and edit content, which covers most of what you will ask the agent to do, without being able to install plugins or change site settings. If something goes wrong, you revoke that one account's access and your admin login is untouched.
Never let the agent publish directly on a site that earns money or carries client work until you have reviewed several of its drafts. Drafts are free to check and free to delete. Published mistakes are not.
What happens after 30 November
This is the question I get most, so let me answer it plainly. After 30 November 2026, the free access for Jetpack sites and WordPress.com free-plan sites ends, and the connection becomes a feature of the plans listed on the pricing pages. Your content does not disappear. Drafts stay drafts, published posts stay published, and images stay where they are. What changes is whether the agent can keep connecting without a paid plan.
Conclusion: a tool worth trying before the window closes
If you run a WordPress site — especially if you build with Elementor and spend your days in the dashboard — connect it while it is free, start read-only, and give it one real week of small jobs. If it saves you an hour, it has earned its place. If it does not, disconnect it and move on. Either way, you will have tried the most practical AI feature WordPress has shipped this year, and you will know, from your own site rather than a headline, whether an AI assistant belongs in your workflow.
Frequently Asked Questions
Is connecting ChatGPT or Claude to WordPress really free right now?
Yes, through November 2026. WordPress.com supports the connection on free plans through November, and Jetpack sites can connect free until 30 November 2026 with no paid Jetpack plan required. Your ChatGPT or Claude plan itself is separate and unchanged.
Do I need a WordPress.com site, or does it work on self-hosted WordPress?
Both. WordPress.com sites connect directly through the ChatGPT plugin or Claude connector. Self-hosted sites connect through Jetpack, which links your site to WordPress.com and enables the same connection.
Can the AI publish posts without me checking them?
Only if you give it that power and ask it to. I strongly recommend keeping everything as drafts until you have reviewed several of them. Start with read-only access, then move to editing with drafts, and only consider direct publishing once you trust the workflow.
Is it safe to give an AI access to my site?
It is as safe as the account you connect. The agent can only do what that WordPress user can do, so use a dedicated editor-level account where possible, keep a tested backup, and revoke access any time by removing the app, connector, or application password.
What happens to my content when the free period ends on 30 November 2026?
Nothing is deleted. Your posts, pages, and media stay exactly as they are. Only the connection itself may require a paid plan afterwards, and WordPress and Jetpack say their pricing pages will show which plans include AI agent access.
Does this work well with Elementor-built pages?
It works best on standard posts and simple pages. For Elementor templates and heavily designed landing pages, I recommend letting the agent prepare text and making layout changes in Elementor yourself, or testing on staging first, so a content edit never surprises you on the front end.
