AI Tools
Unusual Activity Detected on ChatGPT? What to Do Next
An unusual activity alert can reflect sign-in, usage, or network signals. Secure the account first, test a trusted device and network, then send persistent errors to OpenAI Support.
Contents
If ChatGPT says “Unusual Activity Detected,” check where you saw it: during sign-in, inside ChatGPT, or in an email. OpenAI uses suspicious activity alerts for potentially risky sign-ins or usage. The message alone does not prove your account was taken over, and it does not prove the problem is only your browser. As of October 10, 2026, OpenAI has not published a universal time for the alert to clear. Here is a safer order of operations.
What can trigger the alert?
OpenAI lists unexpected devices or locations, sudden changes in usage, and more concurrent sessions than usual as possible signals. Its error guide also mentions traffic that resembles automation. A VPN, proxy, Private Relay, browser extension, or automation tool can complicate troubleshooting. The alert may restrict some features, but the wording does not identify the exact cause. Save the full message and the time it appeared before changing your setup.
Secure the account, then troubleshoot the connection
- Look for access you do not recognize. If you see unfamiliar sessions or account changes, change your OpenAI password and secure the associated email account. Review Active sessions in ChatGPT security settings and sign out unfamiliar devices. If needed, sign out of all sessions; OpenAI says this can take up to 30 minutes to finish across devices.
- Turn on multi-factor authentication. MFA helps protect future sign-ins, but enabling it does not end existing sessions. If you sign in through Google, Apple, Microsoft, or your organization’s SSO, check that identity provider too.
- Try one trusted setup. Disable VPNs, proxies, Private Relay, and browser extensions or automation tools that may interfere. Restart the browser or device. Try a private window on a device and network you normally use. If the alert remains, clear ChatGPT cookies/cache or test another browser or device. Clearing cookies signs you out.
- Keep personal credentials personal. OpenAI’s account sharing policy says an individual account is for the person who created it, although that person may use multiple devices. Sharing credentials adds privacy and security risks and may trigger alerts. Other people should use their own accounts or an appropriate team workspace.
If the warning arrived only by email or appeared on an unfamiliar site, do not enter a password or verification code there. Open chatgpt.com or the OpenAI Help Center yourself and check from there.
How long does it last, and when should you contact Support?
OpenAI does not publish a fixed cooldown for every unusual activity alert. It also does not promise that switching networks or clearing cookies will restore access immediately. After securing the account and trying a clean setup, contact OpenAI Support if the banner or feature restriction persists. You do not need to wait for an unofficial “24-hour” or “seven-day” timer.
Include the account email, a screenshot, the first occurrence with time zone, device/browser/OS details, whether you used a VPN or proxy, and any request, ray, or device ID shown. Never send your password, one-time code, or recovery codes.
Common questions
Does this mean my account is banned?
The alert alone cannot establish that. A clear account deactivation notice is a different issue; use our ChatGPT account deactivation appeal guide if that is the message you actually see.
Will clearing cookies or changing networks always fix it?
No. These steps help rule out local session and network interference. They cannot resolve every account security or review issue.
Can using several devices trigger it?
One person may use their account on multiple devices. OpenAI names unusual concurrent sessions as one possible signal, but publishes no device count that automatically triggers an alert. For sign-in challenges, see our ChatGPT login and verification guide.