Weekly report · 2017-W16

The state of AI crawling

Measured on 30,463 reachable domains for the week ending 2017-04-23. 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.

← 2017-W15  ·  2017-W17 →

30,463
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 11.3% 3,436 of 30,461 new
CCBot Common Crawl Foundation archive 11.1% 3,375 of 30,461 new
Bytespider ByteDance training 10.7% 3,252 of 30,461 new
ClaudeBot Anthropic training 9.9% 3,018 of 30,461 new
Meta-ExternalAgent Meta training 8.9% 2,723 of 30,461 new
Google-Extended Google training 8.9% 2,720 of 30,461 new
Amazonbot Amazon training 8.9% 2,714 of 30,461 new
Applebot-Extended Apple training 8.7% 2,652 of 30,461 new
anthropic-ai Anthropic training 4.2% 1,268 of 30,461 new
omgilibot Webz.io training 4.1% 1,259 of 30,461 new
Diffbot Diffbot training 3.8% 1,170 of 30,461 new
cohere-ai Cohere live retrieval 3.8% 1,151 of 30,461 new
omgili Webz.io training 3.8% 1,144 of 30,461 new
SemrushBot-OCOB Semrush training 3.7% 1,131 of 30,461 new
PerplexityBot Perplexity answer index 3.6% 1,092 of 30,461 new
ChatGPT-User OpenAI live retrieval 3.6% 1,089 of 30,461 new
FacebookBot Meta training 3.2% 983 of 30,461 new
YouBot You.com answer index 2.9% 881 of 30,461 new
ImagesiftBot ImageSift (Hive) training 2.7% 834 of 30,461 new
Timpibot Timpi training 2.7% 832 of 30,461 new
Scrapy Zyte (open-source framework) training 2.4% 723 of 30,461 new
AI2Bot Allen Institute for AI training 2.2% 685 of 30,461 new
OAI-SearchBot OpenAI answer index 2.2% 683 of 30,461 new
cohere-training-data-crawler Cohere training 2.0% 598 of 30,461 new
DuckAssistBot DuckDuckGo live retrieval 1.7% 527 of 30,461 new
Meta-ExternalFetcher Meta live retrieval 1.7% 526 of 30,461 new
PanguBot Huawei training 1.7% 514 of 30,461 new
Claude-SearchBot Anthropic answer index 1.6% 479 of 30,461 new
Claude-User Anthropic live retrieval 1.5% 470 of 30,461 new
Kangaroo Bot Kangaroo LLM training 1.5% 459 of 30,461 new
Perplexity-User Perplexity live retrieval 1.5% 444 of 30,461 new
MistralAI-User Mistral AI live retrieval 1.4% 425 of 30,461 new
Applebot Apple answer index 1.1% 349 of 30,461 new
Google-CloudVertexBot Google live retrieval 1.1% 349 of 30,461 new
Meta-WebIndexer Meta answer index 1.1% 347 of 30,461 new
GoogleOther Google training 1.1% 331 of 30,461 new
TikTokSpider ByteDance live retrieval 0.8% 255 of 30,461 new
Amzn-SearchBot Amazon answer index 0.7% 216 of 30,461 new
Amzn-User Amazon live retrieval 0.4% 126 of 30,461 new
MistralAI-Index Mistral AI answer index 0.2% 69 of 30,461 new
facebookexternalhit Meta live retrieval 0.2% 58 of 30,461 new
MistralAI-Training Mistral AI training 0.2% 57 of 30,461 new
Bingbot Microsoft answer index 0.1% 40 of 30,461 new
Googlebot-News Google answer index 0.1% 39 of 30,461 new
msnbot Microsoft answer index 0.1% 27 of 30,461 new
Meta-ExternalAds Meta training 0.1% 20 of 30,461 new
Googlebot Google answer index 0.1% 16 of 30,461 new
OAI-AdsBot OpenAI live retrieval 0.0% 3 of 30,461 new

Trend share of measured sites

% of sites20151050% of sites08-2209-0109-11block 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=451)17.1%.fr (n=326)16.6%.de (n=837)16%.jp (n=420)8.8%.com (n=14977)6.6%.br (n=380)6.1%.org (n=1686)5.2%.net (n=1320)3.9%.ru (n=1161)3%.in (n=303)2.6%.edu (n=477)2.3%.io (n=657)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, 2017-W16", 2017-04-23. https://crawlcensus.com/report/2017-W16

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