Weekly report · 2002-W36

The state of AI crawling

Measured on 30,357 reachable domains for the week ending 2002-09-08. Every figure is the result of an automated audit run the same way on every domain: read robots.txt, resolve the rules for 48 tracked agents, then send live requests as four of them.

← 2002-W35  ·  2002-W37 →

30,357
domains measured
7%
block an answer engine
19%
refuse AI agents at the edge
14%
publish an llms.txt

Block rate by crawler with change on the previous week

CrawlerOperatorContent used forBlock rateSitesWeek on week
GPTBot OpenAI training 7.7% 2,348 of 30,359 new
CCBot Common Crawl Foundation archive 7.4% 2,254 of 30,359 new
Bytespider ByteDance training 7.2% 2,175 of 30,359 new
ClaudeBot Anthropic training 6.4% 1,954 of 30,359 new
Meta-ExternalAgent Meta training 5.5% 1,669 of 30,359 new
Amazonbot Amazon training 5.5% 1,662 of 30,359 new
Applebot-Extended Apple training 5.5% 1,656 of 30,359 new
Google-Extended Google training 5.4% 1,648 of 30,359 new
anthropic-ai Anthropic training 5.1% 1,551 of 30,359 new
omgilibot Webz.io training 5.1% 1,540 of 30,359 new
Diffbot Diffbot training 4.9% 1,485 of 30,359 new
cohere-ai Cohere live retrieval 4.8% 1,468 of 30,359 new
omgili Webz.io training 4.7% 1,425 of 30,359 new
SemrushBot-OCOB Semrush training 3.5% 1,075 of 30,359 new
PerplexityBot Perplexity answer index 3.4% 1,029 of 30,359 new
ChatGPT-User OpenAI live retrieval 3.3% 1,011 of 30,359 new
FacebookBot Meta training 3.1% 940 of 30,359 new
YouBot You.com answer index 3.0% 898 of 30,359 new
ImagesiftBot ImageSift (Hive) training 2.7% 825 of 30,359 new
Timpibot Timpi training 2.7% 825 of 30,359 new
Scrapy Zyte (open-source framework) training 2.4% 727 of 30,359 new
AI2Bot Allen Institute for AI training 2.3% 706 of 30,359 new
Google-CloudVertexBot Google live retrieval 2.3% 695 of 30,359 new
GoogleOther Google training 2.2% 657 of 30,359 new
OAI-SearchBot OpenAI answer index 2.1% 627 of 30,359 new
cohere-training-data-crawler Cohere training 2.0% 613 of 30,359 new
Meta-ExternalFetcher Meta live retrieval 1.8% 534 of 30,359 new
DuckAssistBot DuckDuckGo live retrieval 1.7% 522 of 30,359 new
PanguBot Huawei training 1.7% 516 of 30,359 new
Claude-User Anthropic live retrieval 1.7% 502 of 30,359 new
Perplexity-User Perplexity live retrieval 1.6% 477 of 30,359 new
Claude-SearchBot Anthropic answer index 1.6% 474 of 30,359 new
MistralAI-User Mistral AI live retrieval 1.5% 467 of 30,359 new
Kangaroo Bot Kangaroo LLM training 1.5% 456 of 30,359 new
MistralAI-Training Mistral AI training 1.5% 446 of 30,359 new
Meta-WebIndexer Meta answer index 1.2% 365 of 30,359 new
Applebot Apple answer index 1.2% 358 of 30,359 new
TikTokSpider ByteDance live retrieval 0.8% 256 of 30,359 new
Amzn-SearchBot Amazon answer index 0.8% 253 of 30,359 new
Amzn-User Amazon live retrieval 0.6% 195 of 30,359 new
MistralAI-Index Mistral AI answer index 0.3% 99 of 30,359 new
facebookexternalhit Meta live retrieval 0.2% 53 of 30,359 new
Googlebot-News Google answer index 0.1% 36 of 30,359 new
Bingbot Microsoft answer index 0.1% 32 of 30,359 new
Meta-ExternalAds Meta training 0.1% 25 of 30,359 new
Googlebot Google answer index 0.1% 17 of 30,359 new
msnbot Microsoft answer index 0.1% 17 of 30,359 new
OAI-AdsBot OpenAI live retrieval 0.0% 13 of 30,359 new

Trend share of measured sites

% of sites20151050% of sites08-2709-2510-25block an answer engineblock a training crawlerpublish llms.txt

Where the blocking is share blocking an answer engine, by top-level domain

bar chart.fr (n=327)17.1%.uk (n=442)16.7%.de (n=832)15.6%.jp (n=407)9.1%.br (n=366)7.7%.com (n=14859)7.3%.org (n=1845)6.1%.net (n=1301)5.8%.in (n=310)3.2%.ru (n=1148)3%.io (n=657)2.6%.edu (n=469)2.6%

Only TLDs with at least 40 measured domains are shown, and each bar carries its sample size. A TLD is a weak proxy for jurisdiction. Full segmentation.

Changes recorded this week 0 total

No crawler policy changes were recorded in this week.

Full change log · RSS

Cite this CC BY 4.0

The figures on this page are frozen to the week shown and will not be edited after publication. Corrections are appended, not overwritten.

Crawl Census, "The state of AI crawling, 2002-W36", 2002-09-08. https://crawlcensus.com/report/2002-W36

Machine-readable: census totals, full corpus, crawler registry. Method: methodology.