Weekly report · 2001-W42

The state of AI crawling

Measured on 30,005 reachable domains for the week ending 2001-10-21. 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.

← 2001-W41  ·  2001-W43 →

30,005
domains measured
9%
block an answer engine
19%
refuse AI agents at the edge
15%
publish an llms.txt

Block rate by crawler with change on the previous week

CrawlerOperatorContent used forBlock rateSitesWeek on week
GPTBot OpenAI training 9.4% 2,822 of 30,005 new
CCBot Common Crawl Foundation archive 9.2% 2,761 of 30,005 new
Bytespider ByteDance training 9.0% 2,695 of 30,005 new
ClaudeBot Anthropic training 8.2% 2,459 of 30,005 new
Meta-ExternalAgent Meta training 7.4% 2,215 of 30,005 new
Amazonbot Amazon training 7.3% 2,176 of 30,005 new
Applebot-Extended Apple training 7.2% 2,164 of 30,005 new
Google-Extended Google training 7.2% 2,149 of 30,005 new
anthropic-ai Anthropic training 6.9% 2,073 of 30,005 new
omgilibot Webz.io training 6.8% 2,048 of 30,005 new
Diffbot Diffbot training 6.7% 2,002 of 30,005 new
cohere-ai Cohere live retrieval 6.7% 1,999 of 30,005 new
omgili Webz.io training 6.5% 1,938 of 30,005 new
Google-CloudVertexBot Google live retrieval 4.1% 1,231 of 30,005 new
GoogleOther Google training 4.0% 1,192 of 30,005 new
SemrushBot-OCOB Semrush training 3.5% 1,056 of 30,005 new
PerplexityBot Perplexity answer index 3.4% 1,020 of 30,005 new
MistralAI-Training Mistral AI training 3.3% 995 of 30,005 new
ChatGPT-User OpenAI live retrieval 3.3% 990 of 30,005 new
FacebookBot Meta training 3.1% 926 of 30,005 new
YouBot You.com answer index 3.0% 888 of 30,005 new
Timpibot Timpi training 2.7% 821 of 30,005 new
ImagesiftBot ImageSift (Hive) training 2.7% 817 of 30,005 new
Scrapy Zyte (open-source framework) training 2.4% 716 of 30,005 new
AI2Bot Allen Institute for AI training 2.3% 699 of 30,005 new
cohere-training-data-crawler Cohere training 2.1% 617 of 30,005 new
OAI-SearchBot OpenAI answer index 2.1% 616 of 30,005 new
Meta-ExternalFetcher Meta live retrieval 1.8% 553 of 30,005 new
PanguBot Huawei training 1.7% 508 of 30,005 new
DuckAssistBot DuckDuckGo live retrieval 1.7% 503 of 30,005 new
Claude-User Anthropic live retrieval 1.7% 502 of 30,005 new
Perplexity-User Perplexity live retrieval 1.6% 479 of 30,005 new
Claude-SearchBot Anthropic answer index 1.6% 466 of 30,005 new
MistralAI-User Mistral AI live retrieval 1.5% 460 of 30,005 new
Kangaroo Bot Kangaroo LLM training 1.5% 441 of 30,005 new
Meta-WebIndexer Meta answer index 1.3% 395 of 30,005 new
Applebot Apple answer index 1.2% 359 of 30,005 new
Amzn-SearchBot Amazon answer index 0.9% 259 of 30,005 new
TikTokSpider ByteDance live retrieval 0.8% 251 of 30,005 new
Amzn-User Amazon live retrieval 0.7% 207 of 30,005 new
MistralAI-Index Mistral AI answer index 0.4% 110 of 30,005 new
facebookexternalhit Meta live retrieval 0.2% 57 of 30,005 new
Googlebot-News Google answer index 0.1% 39 of 30,005 new
Meta-ExternalAds Meta training 0.1% 37 of 30,005 new
Bingbot Microsoft answer index 0.1% 36 of 30,005 new
Googlebot Google answer index 0.1% 19 of 30,005 new
msnbot Microsoft answer index 0.1% 19 of 30,005 new
OAI-AdsBot OpenAI live retrieval 0.0% 13 of 30,005 new

Trend share of measured sites

% of sites20151050% of sites09-0310-0211-01block an answer engineblock a training crawlerpublish llms.txt

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

bar chart.uk (n=428)17.5%.fr (n=323)16.4%.de (n=821)16.3%.com (n=14650)9.4%.jp (n=419)9.1%.org (n=1830)9%.net (n=1298)8.6%.br (n=360)6.9%.in (n=304)4.6%.io (n=656)3.5%.ru (n=1151)2.9%.edu (n=472)2.8%

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, 2001-W42", 2001-10-21. https://crawlcensus.com/report/2001-W42

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