X automated account label: who needs it, how to add it

#Quick answer

The X automated account label is a self-selected profile label that says an account is not human-run, and it only works in a pair: the bot carries the label, and a human-run "managing account" is shown as the operator. X's Developer Guidelines make it a requirement for automated accounts using the X API, together with a bio that states the account is a bot and who runs it. The setting lives on the bot account under Settings and privacy, Your account, Account information, Automation. X's own help page still calls the feature "currently in testing", which is why the menu is missing for a lot of accounts.

Last updated: October 2026

#TL;DR

The label is free, takes a minute, and is the one piece of bot compliance on X (formerly Twitter) that is visible to every person who reads one of your posts. It is also the piece with the worst documentation: the help page describing it was written for a 2021 test group and never updated, the path to the setting is not published anywhere on help.x.com, and the single most common complaint about it on X's developer forum is that the menu does not exist or will not switch off.

What follows is what the label actually asserts, who X says needs one, the path people report using, what to do when the menu misbehaves, and the rules that keep applying after the label is on.

#What the label asserts, and what it does not

X's profile labels page groups the identity signals on a profile into ones X applies (checkmarks, affiliation badges) and ones you select yourself. The automated label sits in the self-selected group, next to the parody, commentary and fan labels.

Read the claim literally, because it is narrower than it looks. X's help page says that when an account displays the automated label, "you know the account is generating automated content not produced by a human". Two consequences follow:

• It is a statement about who composes the content, not about which tool published it. A person who writes every post themselves and schedules them does not become an automated account.
• It is a statement about the account, not about an individual post. There is no per-post version of this label.

Half the mechanism gets skipped in most summaries. The label does not stand alone: X requires the bot to stay connected to a human-run account, and the connection is what the reader sees. That is the accountability design. A labelled bot with no identifiable operator was never the intended end state.

#Who X says needs one

The requirement is not in the help centre. It is in the Developer Guidelines, which set out what an automated account that posts, replies or interacts on behalf of users has to do. Accessed on 4 October 2026, that page requires an automated account to:

1. Enable the "Automated" profile label, which appears under the bot's name and handle.
2. State clearly in the bio that it is a bot and who operates it. X's own examples are "Bot by @yourcompany" and "Automated account managed by Example Inc."
3. Be associated with a human-managed account for accountability and contact.
4. Honour opt-outs. If a user says stop, stop, with keyword detection for common opt-out phrases.
5. Use the API. The same page says no scraping, browser automation, or unofficial methods, and that violations result in permanent suspension.

Point 5 is the one to read twice if you are shopping for tooling, and we say the same thing on our own product pages: the API route is the sanctioned one, and anything driving the website instead is a gray area that X's automation rules address directly. Those rules, updated April 2026, list "use non-API-based forms of automation, such as scripting the X website" under Don't, and say the technique may result in permanent suspension of your account. No label changes that.

If you are weighing the sanctioned path, the current cost of entry is covered in X API pricing and the signup mechanics in how to get an X API key.

#How to add the automated account label

X does not publish the click path on the label's help page, which is a genuine documentation gap rather than an oversight you can work around. The path reported consistently by developers, and the one the menu still uses, runs on the bot account:

1. Open Settings and privacy.
2. Go to Your account.
3. Open Account information. You may be asked for your password here.
4. Select Automation.
5. Choose Managing account, and enter the handle of the human-run account that operates the bot.
6. Log in as that managing account to confirm the connection.

Step 6 is the gate. You need working access to both accounts, because the bot cannot nominate an operator the operator has not agreed to. If the managing account has two-factor authentication on, keep that device nearby.

Then do the part people forget: edit the bot's bio so it names the operator in plain language. The label and the bio are separate requirements in the Developer Guidelines, and the bio is the one that survives a screenshot.

#When the menu is missing, or the label will not come off

This is the real state of the feature in 2026, and no other page covering the topic says it plainly.

X's profile labels page describes automated account labels as "currently in testing". The dedicated help page opens with "We're launching a labeling feature", language from the original September 2021 test, and describes the label appearing "once an account accepts the invitation to our test group". X's developer account said in February 2022 that labels were available to all developer-created bot accounts. Both framings are still live on X's properties at once.

Practical symptoms, all reported in long-running threads on X's developer forum:

• No Automation entry under Account information on the bot account, while the same menu shows it on a main account.
• A label that disappeared after a managing account was connected and never came back.
• A disconnect that does not stick: turn off account automation, confirm, refresh, and the label is still there. One thread tracks this across roughly two years.
• "Oops, something went wrong" when opening the Automation menu at all.

If you are stuck, your options are limited and worth being honest about. Try the web interface at x.com rather than the apps, since this menu has historically been web-first. Confirm the managing account is in good standing and not itself labelled. Then file through the support route, which is slow for everything that is not account access, as we cover in how to contact X support. There is no public timeline for a fix, and the test-group framing means X has never committed to one.

Document your attempt. If the label will not enable, a bio that clearly states the account is automated and who runs it is the part you control, and it addresses the same transparency requirement in the most durable place on the profile.

#The label is not a permission slip

The most expensive misreading of the label is treating it as cover. Carrying it does not widen what your bot may do. X's automation rules, updated April 2026, are explicit about the actions that stay off limits regardless:

That AI reply bot line is the one that catches teams building in 2026, because it reads as a green light for the category and then requires written approval per deployment. The reply-automation limits and the per-day ceilings that sit underneath them are covered in X daily limits, and the wider question of which bots survive on the platform in X bots explained.

#If your account is human-run and you use a tool

This is the case most people searching for the label are actually in, and the answer is different.

An account where a person decides what gets said, approves it, and answers for it is not an automated account under X's definition, even if software helps draft or publish. The label is for accounts "not human-run". Applying it to your own primary account because you use a scheduler would be inaccurate in the direction that costs you credibility with readers.

What does change is your exposure. Cloud tools hold your session on their servers and post from shared IP pools, which is the pattern platform integrity systems are built to spot. X-Autopilot drafts posts and replies in your voice and runs them through real Chrome on your own Mac, so the session cookie stays on your machine, the traffic comes from your residential IP, and no delegate account is involved. That is a lower detection surface than a shared cloud pipeline, not an exemption. Browser automation remains outside what X's automation rules sanction, you stay responsible for what your account does, and we will not tell you otherwise.

The honest split: if you are building a bot, use the API, label it, name the operator in the bio, and keep the behaviour inside the automation rules. If you are a person who wants to post more without living in the app, you are not the account the label was designed for.