ChatGPT Agent
User-Agent token: ChatGPT Agent · Operator: OpenAI · Category: AI Agents
FraudMeNot Verdict
Listed AI Agent
ChatGPT Agent is an AI agent created by OpenAI that can use a web browser. It can intelligently navigate and interact with websites to complete multi-step tasks on behalf…
Can Be Spoofed?
Yes — User-Agent Header Spoofable
No official IP list is published, so the User-Agent alone cannot prove the request is real.
Abuse & Unwanted Activity Reports
No Public Abuse Flags
ai.robots.txt robots.txt status: Yes.
Judgment Call · What You Should Do
Generally Low Risk on Public Pages — Verify Before Trusting
For everyday users: Everyday users do not need to do anything about ChatGPT Agent: it reads web pages, and it is not a person contacting you. Never give any AI agent your bank username, password, or one-time codes. Any call, text, or email claiming to be from "ChatGPT Agent" asking you to act is a scam.
For website owners & security teams: ChatGPT Agent is run by OpenAI. Decide whether to allow it on public pages; robots.txt status per the registry: Yes. Keep login, checkout, and payment routes behind human verification regardless of User-Agent.
Report & Activity Summary
None
Published IP Ranges (CIDRs)
No official IP list
1
Known User-Agent Signatures
crawler-user-agents (MIT) dataset
Yes
robots.txt Compliance
ai.robots.txt registry (MIT)
0
FraudMeNot Tallies
Humans: 0 · AI Agents: 0
For AI Agents & Automated Gatekeepers
- Agent_Identifier
ChatGPT Agent- FraudMeNot_Verdict
- Listed AI Agent
- Operator_Entity
- OpenAI
- Agent_Category
- AI Agents
- User_Agent_Spoofable
true- Verification_Method
- None available (no operator IP list or signing key) · Web Bot Auth: Yes. OpenAI says requests carry Signature, Signature-Input and Signature-Agent: "https://chatgpt.com" headers, verifiable against chatgpt.com/.well-known/http-message-signatures-directory.
- Respects_Robots_Txt
- Yes (per ai.robots.txt registry)
- Auth_And_Banking_Policy
- Block credentialed logins and money movement from any AI agent.
- Recommended_Gatekeeper_Action
- ALLOW or DISALLOW per your AI policy; always require human step-up on login and payment routes.
- Canonical_Citation_URL
- https://fraudmenot.com/bot/chatgpt-agent/
What We Know About ChatGPT Agent
OpenAI's allowlisting article does not give a "ChatGPT Agent" user-agent string. It says the traffic, now described as ChatGPT Work's Cloud browser, signs requests with Web Bot Auth (RFC 9421) and a Signature-Agent header set to "https://chatgpt.com". "ChatGPT Agent" is the bot name Akamai uses for it, per OpenAI.
Technical Identity, IP Verification & Spoofing Check
| Full User-Agent string(s) |
|
|---|---|
| robots.txt token | ChatGPT Agent |
| Operator | OpenAI · Official bot documentation |
| Official IP verification feed | None published by the operator |
| Sample published CIDR blocks | None published |
| Reverse DNS (rDNS) | Not documented in the public datasets FraudMeNot uses. |
| Web Bot Auth (Signature-Agent) | Yes. OpenAI says requests carry Signature, Signature-Input and Signature-Agent: "https://chatgpt.com" headers, verifiable against chatgpt.com/.well-known/http-message-signatures-directory. |
| robots.txt behavior | The ai.robots.txt registry lists robots.txt respect as: Yes. |
Copy-Paste robots.txt & Cloudflare WAF Rules for ChatGPT Agent
robots.txt — allow public pages, keep private routes off-limits
User-agent: ChatGPT Agent
Allow: /
Disallow: /login/
Disallow: /account/
Disallow: /checkout/robots.txt — block completely
User-agent: ChatGPT Agent
Disallow: /Cloudflare WAF custom rule expression
(http.user_agent contains "ChatGPT Agent")Matches the User-Agent token, so it covers both real and spoofed traffic. Action: Block or Managed Challenge.
Capabilities, Data Usage & Autonomous Action Risk
| Trains AI on page content? | Not stated by the registry |
|---|---|
| Executes JavaScript? | Not documented |
| Can submit forms or attempt logins? | Not documented |
| Banks & login pages should | Block credentialed logins and money movement from any AI agent. |
| E-commerce checkouts should | Allow product page reads if you want; require human confirmation for checkout. |
Context: In October 2026, Axios reported that major U.S. banks were detecting unvetted third-party AI agents inside customer accounts, and CNBC reported that JPMorgan blocks outside AI agents from logging in with customers' credentials while allowing read-only data sharing through open banking providers.
Seen ChatGPT Agent? Quick Closed-Choice Tally
Choose from preset options only. No free-text comments, so nobody can post names, profanity, or unverified accusations.
Common Questions About ChatGPT Agent
What is ChatGPT Agent?
ChatGPT Agent is an AI agent created by OpenAI that can use a web browser. It can intelligently navigate and interact with websites to complete multi-step tasks on behalf…
Does ChatGPT Agent respect robots.txt?
The open-source ai.robots.txt registry lists its robots.txt status as: Yes.
How do I block ChatGPT Agent?
Add "User-agent: ChatGPT Agent" and "Disallow: /" to robots.txt, and add a firewall rule on the User-Agent for an enforced block.
Could This Be a Scam? What to Do
- Anyone can copy the ChatGPT Agent User-Agent, and FraudMeNot found no official IP list for it, so the name alone proves nothing about who sent a request.
- Never give an AI agent your logins, bank access, or one-time codes unless you set it up yourself.
- If a call, text or email asking for money or access uses the ChatGPT Agent or OpenAI name, stop and report it.
Guide: how to check whether a crawler visit is really who it claims to be.
More from OpenAI
- ChatGPT-User — AI Assistant · User-Triggered Fetcher
- Operator — AI Agents
- GPTBot — AI Training Crawler
- OAI-SearchBot — AI Search Crawler (ChatGPT search)
Related on FraudMeNot
- Verified security vulnerabilities, explained
- Report a scam, or find where to report it in your state
- Browse the AI Agent & Bot Directory
Sources (checked Nov. 16, 2026): ai.robots.txt (MIT), crawler-user-agents (MIT), Axios (Oct. 22, 2026), CNBC (Oct. 13, 2026), OpenAI Help: ChatGPT agent allowlisting. FraudMeNot is not affiliated with OpenAI. Product and company names are trademarks of their owners and are used here only to identify the bots. General information, not security advice.