Rattlesnakes By Mail

Bingbot vs GPTBot

Bingbot and GPTBot compared on 12 documented fields, from 19 claims, last verified 2026-09-14.

Field	Bingbot	GPTBot	Comparison
cites_sources	yes	not documented	different
follows_noarchive	yes	not documented	different
honours_crawl_delay	yes	not_documented	different
opt_out_mechanism	yes	not documented	different
publishes_ip_list	yes	yes	same
purpose_documented	yes	training	different
reads_llms_txt	not_documented	not documented	different
respects_robots_txt	yes	yes	same
robots_txt_fetch_marker	not documented	yes	different
separate_search_and_training_tokens	no	yes	different
user_agent_string_full	Mozilla/5.0 (compatible; bingbot/2.0; +http://www.bing.com/bingbot.htm)	Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.4; +https://openai.com/gptbot	different
verifiable_by_rdns	yes	not_documented	different

Differences

Field	Bingbot statement	GPTBot statement
cites_sources	Copilot Search in Bing delivers clearly cited sources alongside generative answers to support publishers and content owners.	not documented
follows_noarchive	Bingbot excludes noarchive-tagged content from Copilot and Chat answers and from generative AI training data.	not documented
honours_crawl_delay	Microsoft states that a crawl-delay directive in robots.txt always takes precedence over Bing Webmaster Tools crawl control settings for Bingbot.	OpenAI does not publish whether GPTBot honours the robots.txt Crawl-delay directive.
opt_out_mechanism	Microsoft lets publishers block Bingbot content from Copilot answers and AI training using noarchive and nocache tags.	not documented
purpose_documented	Bingbot serves as Microsoft's standard crawler and handles most of Bing's daily web crawling.	GPTBot crawls content that may be used to train OpenAI generative AI foundation models.
reads_llms_txt	Microsoft does not publish whether Bingbot reads llms.txt files.	not documented
robots_txt_fetch_marker	not documented	OpenAI may add a robots.txt marker to the GPTBot user agent string when GPTBot fetches robots.txt files.
separate_search_and_training_tokens	Microsoft controls Bingbot's use in AI training through meta tags rather than publishing a separate training-specific crawler token.	OpenAI separates search and training controls into two robots.txt tokens: OAI-SearchBot for search and GPTBot for training.
user_agent_string_full	Bingbot sends a user agent string containing bingbot/2.0 and a link to Microsoft's bingbot.htm page.	GPTBot sends the user agent string Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.4; +https://openai.com/gptbot.
verifiable_by_rdns	Microsoft verifies Bingbot traffic through reverse DNS lookups that resolve to hostnames ending in search.msn.com.	OpenAI does not publish a reverse-DNS verification method for GPTBot.

Sources