Weekly report · 2002-W07

The state of AI crawling

Measured on 30,051 reachable domains for the week ending 2002-02-17. 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-W06  ·  2002-W08 →

30,051
domains measured
8%
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 8.3% 2,509 of 30,049 new
CCBot Common Crawl Foundation archive 8.1% 2,431 of 30,049 new
Bytespider ByteDance training 7.8% 2,358 of 30,049 new
ClaudeBot Anthropic training 7.1% 2,124 of 30,049 new
Meta-ExternalAgent Meta training 6.2% 1,864 of 30,049 new
Applebot-Extended Apple training 6.1% 1,830 of 30,049 new
Amazonbot Amazon training 6.1% 1,827 of 30,049 new
Google-Extended Google training 6.1% 1,824 of 30,049 new
anthropic-ai Anthropic training 5.8% 1,743 of 30,049 new
omgilibot Webz.io training 5.7% 1,717 of 30,049 new
Diffbot Diffbot training 5.6% 1,671 of 30,049 new
cohere-ai Cohere live retrieval 5.5% 1,657 of 30,049 new
omgili Webz.io training 5.3% 1,606 of 30,049 new
SemrushBot-OCOB Semrush training 3.5% 1,056 of 30,049 new
PerplexityBot Perplexity answer index 3.4% 1,028 of 30,049 new
ChatGPT-User OpenAI live retrieval 3.3% 1,004 of 30,049 new
FacebookBot Meta training 3.1% 940 of 30,049 new
YouBot You.com answer index 3.0% 897 of 30,049 new
Google-CloudVertexBot Google live retrieval 2.9% 885 of 30,049 new
GoogleOther Google training 2.8% 845 of 30,049 new
Timpibot Timpi training 2.8% 829 of 30,049 new
ImagesiftBot ImageSift (Hive) training 2.7% 823 of 30,049 new
Scrapy Zyte (open-source framework) training 2.4% 727 of 30,049 new
AI2Bot Allen Institute for AI training 2.3% 697 of 30,049 new
MistralAI-Training Mistral AI training 2.1% 640 of 30,049 new
OAI-SearchBot OpenAI answer index 2.1% 619 of 30,049 new
cohere-training-data-crawler Cohere training 2.0% 612 of 30,049 new
Meta-ExternalFetcher Meta live retrieval 1.8% 541 of 30,049 new
PanguBot Huawei training 1.7% 516 of 30,049 new
DuckAssistBot DuckDuckGo live retrieval 1.7% 511 of 30,049 new
Claude-User Anthropic live retrieval 1.6% 492 of 30,049 new
Perplexity-User Perplexity live retrieval 1.6% 471 of 30,049 new
Claude-SearchBot Anthropic answer index 1.5% 461 of 30,049 new
MistralAI-User Mistral AI live retrieval 1.5% 455 of 30,049 new
Kangaroo Bot Kangaroo LLM training 1.5% 449 of 30,049 new
Meta-WebIndexer Meta answer index 1.3% 382 of 30,049 new
Applebot Apple answer index 1.2% 360 of 30,049 new
Amzn-SearchBot Amazon answer index 0.8% 254 of 30,049 new
TikTokSpider ByteDance live retrieval 0.8% 252 of 30,049 new
Amzn-User Amazon live retrieval 0.7% 200 of 30,049 new
MistralAI-Index Mistral AI answer index 0.3% 105 of 30,049 new
facebookexternalhit Meta live retrieval 0.2% 57 of 30,049 new
Googlebot-News Google answer index 0.1% 38 of 30,049 new
Bingbot Microsoft answer index 0.1% 36 of 30,049 new
Meta-ExternalAds Meta training 0.1% 32 of 30,049 new
msnbot Microsoft answer index 0.1% 19 of 30,049 new
Googlebot Google answer index 0.1% 18 of 30,049 new
OAI-AdsBot OpenAI live retrieval 0.0% 12 of 30,049 new

Trend share of measured sites

% of sites20151050% of sites08-3109-2910-29block 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=426)17.4%.fr (n=321)16.8%.de (n=820)16.3%.jp (n=417)8.9%.com (n=14679)8.2%.org (n=1836)7.2%.net (n=1297)6.9%.br (n=360)6.4%.in (n=301)3.3%.io (n=655)3.1%.ru (n=1153)2.9%.edu (n=473)2.3%

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-W07", 2002-02-17. https://crawlcensus.com/report/2002-W07

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