“`html
Permanently banned by OpenAI for “Cyber Abuse”: My Experience
I have been a $20/month Plus subscriber for over two years, using ChatGPT daily for work and dev projects. Last week, my account was deactivated for “Cyber Abuse” (per GPT TOS is phishing/malware, etc.). I have never done anything remotely close to that, but the way OpenAI handled it has been a total nightmare.
OpenAI’s entire warning system relies on emails that their own documentation admits often hit spam filters.
Timeline: All notifications in spam folder
- May 7 – May 12: OpenAI sends 3 separate emails: “Usage Policy Violation & Deactivation Warning”. All go straight to Outlook spam. I never see them.
- May 13: “OpenAI – Access Deactivated” email also goes to spam. I find out I’m locked when I get forced out of all GPT sessions and can’t log back in…
You do not have an account because it has been deleted or deactivated. If you believe this was an error, please contact us through our help center at help.openai.com.
May 13:
- I submit a appeal via web form and within ~1 hour, “After carefully reviewing your account, we are upholding our decision to deactivate your access. We will no longer consider additional requests to appeal this case.” is received.
Was I compromised?
- I was running OpenClaw via Codex OAuth on a new VPS. I was still hardening it, so it was a legit attack surface. My OAuth token was hijacked around May 7.
- Instead of detecting the breach and locking my account for my protection, OpenAI just sent silent emails. They treated a loyal customer like a malicious hacker and nuked a 2-year-old account without a second thought.
Support nightmare (no surprise)
- The first response I got after submitting my appeal from AI-assisted support told me to submit another appeal with more details, despite the fact that the first appeal had already been closed as final.
- A follow-up from a human said: “We kindly recommend referring to the previous email sent to you, as it contains the relevant details.” That email said: “Cyber Abuse.” That’s it. No specifics. No examples. No evidence.
- When I asked for more details about the violation, prompts, IP/geolocation so I could investigate if I’d been hacked, I was told: “For security and privacy reasons, we have limitations on the specific account details and internal security information that we can disclose.”
What I’m asking for:
- A real human to review! Over 2 years of paying for Plus should at least buy me a look from a real person, not AI support slop.
- Block prompts or in app violation notifications, not emails!! – If you suspect a breach, lock the account to protect the user, don’t deactivate!
(Happy to drop screenshots of the emails/rejections in the comments if anyone wants to see the “support” loop in action.)
Key Takeaways
- No in-app warnings or blocked prompts despite multiple attempts.
- Silent emails sent as a warning mechanism, which often hit spam filters.
- No real human review of my appeal; only AI-assisted responses provided.
“`
This HTML document is structured to reflect the content and structure of the original Reddit post, maintaining key facts and the narrative while rewriting in British English.
Originally published at reddit.com. Curated by AI Maestro.
Stay ahead of AI. Get the most important stories delivered to your inbox — no spam, no noise.