Weekly report · 2015-W09

The state of AI crawling

Measured on 30,283 reachable domains for the week ending 2015-03-01. 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.

← 2015-W08  ·  2015-W10 →

30,283
domains measured
6%
block an answer engine
19%
refuse AI agents at the edge
13%
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% 2,502 of 30,280 new
CCBot Common Crawl Foundation archive 8.1% 2,448 of 30,280 new
Bytespider ByteDance training 7.6% 2,308 of 30,280 new
ClaudeBot Anthropic training 6.8% 2,070 of 30,280 new
Amazonbot Amazon training 5.9% 1,786 of 30,280 new
Meta-ExternalAgent Meta training 5.9% 1,785 of 30,280 new
Google-Extended Google training 5.8% 1,752 of 30,280 new
Applebot-Extended Apple training 5.7% 1,724 of 30,280 new
anthropic-ai Anthropic training 4.2% 1,269 of 30,280 new
omgilibot Webz.io training 4.1% 1,241 of 30,280 new
cohere-ai Cohere live retrieval 3.9% 1,181 of 30,280 new
Diffbot Diffbot training 3.8% 1,154 of 30,280 new
SemrushBot-OCOB Semrush training 3.8% 1,150 of 30,280 new
omgili Webz.io training 3.7% 1,125 of 30,280 new
PerplexityBot Perplexity answer index 3.5% 1,063 of 30,280 new
ChatGPT-User OpenAI live retrieval 3.5% 1,050 of 30,280 new
FacebookBot Meta training 3.2% 961 of 30,280 new
YouBot You.com answer index 2.9% 879 of 30,280 new
ImagesiftBot ImageSift (Hive) training 2.8% 833 of 30,280 new
Timpibot Timpi training 2.7% 814 of 30,280 new
Scrapy Zyte (open-source framework) training 2.4% 723 of 30,280 new
AI2Bot Allen Institute for AI training 2.2% 665 of 30,280 new
OAI-SearchBot OpenAI answer index 2.1% 650 of 30,280 new
cohere-training-data-crawler Cohere training 2.0% 607 of 30,280 new
Meta-ExternalFetcher Meta live retrieval 1.8% 546 of 30,280 new
DuckAssistBot DuckDuckGo live retrieval 1.7% 514 of 30,280 new
PanguBot Huawei training 1.7% 500 of 30,280 new
Claude-SearchBot Anthropic answer index 1.6% 478 of 30,280 new
Claude-User Anthropic live retrieval 1.6% 475 of 30,280 new
Kangaroo Bot Kangaroo LLM training 1.5% 455 of 30,280 new
Perplexity-User Perplexity live retrieval 1.5% 448 of 30,280 new
MistralAI-User Mistral AI live retrieval 1.4% 430 of 30,280 new
Google-CloudVertexBot Google live retrieval 1.2% 375 of 30,280 new
Meta-WebIndexer Meta answer index 1.2% 359 of 30,280 new
GoogleOther Google training 1.1% 348 of 30,280 new
Applebot Apple answer index 1.1% 331 of 30,280 new
TikTokSpider ByteDance live retrieval 0.8% 257 of 30,280 new
Amzn-SearchBot Amazon answer index 0.7% 216 of 30,280 new
Amzn-User Amazon live retrieval 0.5% 147 of 30,280 new
MistralAI-Training Mistral AI training 0.3% 78 of 30,280 new
MistralAI-Index Mistral AI answer index 0.2% 74 of 30,280 new
facebookexternalhit Meta live retrieval 0.2% 55 of 30,280 new
Bingbot Microsoft answer index 0.1% 39 of 30,280 new
Googlebot-News Google answer index 0.1% 37 of 30,280 new
msnbot Microsoft answer index 0.1% 26 of 30,280 new
Meta-ExternalAds Meta training 0.1% 20 of 30,280 new
Googlebot Google answer index 0.1% 16 of 30,280 new
OAI-AdsBot OpenAI live retrieval 0.0% 6 of 30,280 new

Trend share of measured sites

% of sites20151050% of sites08-2209-0409-18block 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=448)17.4%.fr (n=328)16.8%.de (n=835)16.5%.jp (n=409)9%.com (n=14838)6.5%.org (n=1671)5.6%.br (n=376)5.3%.net (n=1312)4%.ru (n=1175)3.1%.in (n=300)3%.edu (n=474)2.5%.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, 2015-W09", 2015-03-01. https://crawlcensus.com/report/2015-W09

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