Weekly report · 2012-W42

The state of AI crawling

Measured on 30,044 reachable domains for the week ending 2012-10-21. 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.

← 2012-W41  ·  2012-W43 →

30,044
domains measured
6%
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 6.9% 2,065 of 30,043 new
CCBot Common Crawl Foundation archive 6.8% 2,031 of 30,043 new
Bytespider ByteDance training 6.3% 1,898 of 30,043 new
ClaudeBot Anthropic training 5.5% 1,664 of 30,043 new
Meta-ExternalAgent Meta training 4.5% 1,364 of 30,043 new
Amazonbot Amazon training 4.5% 1,359 of 30,043 new
Google-Extended Google training 4.5% 1,339 of 30,043 new
Applebot-Extended Apple training 4.3% 1,304 of 30,043 new
anthropic-ai Anthropic training 4.3% 1,280 of 30,043 new
omgilibot Webz.io training 4.2% 1,263 of 30,043 new
cohere-ai Cohere live retrieval 3.9% 1,179 of 30,043 new
Diffbot Diffbot training 3.9% 1,175 of 30,043 new
omgili Webz.io training 3.8% 1,148 of 30,043 new
SemrushBot-OCOB Semrush training 3.7% 1,106 of 30,043 new
PerplexityBot Perplexity answer index 3.5% 1,050 of 30,043 new
ChatGPT-User OpenAI live retrieval 3.5% 1,037 of 30,043 new
FacebookBot Meta training 3.2% 952 of 30,043 new
YouBot You.com answer index 2.9% 875 of 30,043 new
ImagesiftBot ImageSift (Hive) training 2.7% 820 of 30,043 new
Timpibot Timpi training 2.7% 806 of 30,043 new
Scrapy Zyte (open-source framework) training 2.4% 720 of 30,043 new
AI2Bot Allen Institute for AI training 2.2% 661 of 30,043 new
OAI-SearchBot OpenAI answer index 2.1% 625 of 30,043 new
cohere-training-data-crawler Cohere training 2.0% 595 of 30,043 new
Meta-ExternalFetcher Meta live retrieval 1.8% 534 of 30,043 new
DuckAssistBot DuckDuckGo live retrieval 1.6% 495 of 30,043 new
PanguBot Huawei training 1.6% 487 of 30,043 new
Claude-User Anthropic live retrieval 1.6% 471 of 30,043 new
Claude-SearchBot Anthropic answer index 1.5% 464 of 30,043 new
Kangaroo Bot Kangaroo LLM training 1.5% 446 of 30,043 new
Perplexity-User Perplexity live retrieval 1.5% 443 of 30,043 new
MistralAI-User Mistral AI live retrieval 1.4% 428 of 30,043 new
Google-CloudVertexBot Google live retrieval 1.3% 380 of 30,043 new
GoogleOther Google training 1.2% 349 of 30,043 new
Meta-WebIndexer Meta answer index 1.2% 346 of 30,043 new
Applebot Apple answer index 1.1% 333 of 30,043 new
TikTokSpider ByteDance live retrieval 0.8% 246 of 30,043 new
Amzn-SearchBot Amazon answer index 0.7% 216 of 30,043 new
Amzn-User Amazon live retrieval 0.5% 149 of 30,043 new
MistralAI-Training Mistral AI training 0.3% 104 of 30,043 new
MistralAI-Index Mistral AI answer index 0.2% 71 of 30,043 new
facebookexternalhit Meta live retrieval 0.2% 52 of 30,043 new
Bingbot Microsoft answer index 0.1% 37 of 30,043 new
Googlebot-News Google answer index 0.1% 37 of 30,043 new
Meta-ExternalAds Meta training 0.1% 22 of 30,043 new
msnbot Microsoft answer index 0.1% 21 of 30,043 new
Googlebot Google answer index 0.1% 16 of 30,043 new
OAI-AdsBot OpenAI live retrieval 0.0% 9 of 30,043 new

Trend share of measured sites

% of sites20151050% of sites08-2209-0809-25block 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=446)17%.fr (n=323)16.7%.de (n=815)15.3%.jp (n=408)8.3%.com (n=14737)6.7%.org (n=1660)5.7%.br (n=363)5.5%.net (n=1303)4.1%.ru (n=1169)3.1%.in (n=315)2.9%.edu (n=471)2.3%.io (n=658)2%

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, 2012-W42", 2012-10-21. https://crawlcensus.com/report/2012-W42

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