ipapi.gr

Bot signals — AI crawler & agent

GPTBot IP addresses

OpenAI GPTBot

Documented GPTBot training crawler egress.

risk.is_airisk.is_crawler Ranges Reviewed
Official documentation (opens in a new tab) All AI crawlers & agents

Check an IP

Is this IP really GPTBot?

Checks the address against what OpenAI GPTBot publishes, the same evidence behind risk.is_ai.

01 — Verification

How IPAPI.GR verifies GPTBot.

Only from what OpenAI GPTBot publishes about itself. No third-party blocklists, no behavioural guesses. Live address counts change with every publication, so check the official file for the current list.

OpenAI GPTBot

risk.is_airisk.is_crawler
Method
Ranges Published IP ranges
Official file
Format
JSON range file
Refresh
Every hour. If a refresh fails, entries expire after 24 hours instead of being trusted forever.

02 — In the response

What a match looks like.

A lookup of a GPTBot address sets risk.is_ai and risk.is_crawler. The flags come with geolocation and network data in the same response, for Pro API keys.

This publication is AI crawling or retrieval traffic, so it counts as both an AI source and a crawler.

The response says which kind of bot an address belongs to, not which operator. Pair it with the user agent: a request that claims to be GPTBot but doesn't match is unverified.

GET /api/lookup/{ip}?key=PRO_KEY
{
  // … location and network fields …
  "risk": {
    "is_ai": true,
    "is_crawler": true,
    "is_scanner": false
  }
}

03 — Configuration

Allow, block or verify it yourself.

robots.txt
User-agent: GPTBot
Disallow: /
OpenAI documents disallowing GPTBot in robots.txt as the opt-out from training.
nginx + jq
# 1. Build a geo include from the official file (cron, e.g. hourly).
#    Keep the old file if the download fails or comes back empty.
curl -fsS https://openai.com/gptbot.json \
  | jq -r '.prefixes[] | (.ipv4Prefix // .ipv6Prefix // empty) | "\(.) 1;"' \
  > /etc/nginx/gptbot-ranges.conf.new && [ -s /etc/nginx/gptbot-ranges.conf.new ] && mv /etc/nginx/gptbot-ranges.conf.new /etc/nginx/gptbot-ranges.conf \
  && nginx -t -q && nginx -s reload

# 2. In the http block: 1 when the client IP is in the published ranges.
geo $gptbot_ip {
    default 0;
    include /etc/nginx/gptbot-ranges.conf;
}

# 3. A request that claims to be GPTBot from any other address is an impostor.
map "$http_user_agent:$gptbot_ip" $fake_gptbot {
    default 0;
    "~*(GPTBot).*:0$" 1;
}

# In the server block:
if ($fake_gptbot) { return 403; }
Built from openai.com/gptbot.json, a JSON file with a prefixes array.

04 — Questions

About GPTBot.

Where does GPTBot publish its IP addresses?

OpenAI GPTBot publishes its addresses at https://openai.com/gptbot.json. IPAPI.GR reads that publication directly, not a third-party copy.

How often is the GPTBot list refreshed?

IPAPI.GR re-fetches it every hour. If a refresh fails, the previous entries expire after at most 24 hours instead of being trusted forever.

Which risk flag does GPTBot set?

A matching address sets risk.is_ai and risk.is_crawler in the lookup response. The risk flags are returned for Pro API keys.

From the blog

Go deeper.

Related operators

Every operator on the coverage page →

Verify GPTBot on every request.

Pro: 40,000,000 requests a day with is_ai, is_crawler and is_scanner in every lookup.

See Pro Start free