Which sites block cohere-training-data-crawler?

Cohere · declared purpose: model training · latest sealed date 2026-08-07.

On 2026-08-07, 1,046 of the 61,552 panel domains that served a parseable robots.txt name cohere-training-data-crawler explicitly. 910 of them (1.5% of parsed) disallow it from the entire site, 108 restrict it to part of the site, and 28 name it and allow it. The remaining 60,506 parsed domains do not mention it at all — their wildcard group governs it.

What the denominator is. On 2026-08-07 we attempted /robots.txt on 100,000 panel domains. 24,130 could not be reached at all (DNS failure, TLS failure, timeout, refused connection) and 14,318 answered with something other than a 200. Both of those groups have no observed AI-crawler policy, and neither is counted as allowing anything — an unreachable site is UNKNOWN, not permissive. Every crawler figure below is out of the 61,552 domains that returned a robots.txt we could parse on that date.

Blocks outright — the file has a group for that exact product token containing Disallow: / with no overriding allow. Partly restricts — a group exists with at least one non-empty Disallow, but not the whole site. Names & allows — a group exists with no effective Disallow. Everything else is not mentioned: the wildcard * group governs it, and wildcard effects are counted separately rather than folded in, because a site that blocks everything has not made a decision about any AI crawler in particular. Matching is exact and case-insensitive per RFC 9309 — never substring.

Get told when cohere-training-data-crawler change their AI crawler rules

Sites publish today's robots.txt and overwrite it — no archive, no history, no notification. We keep a dated daily copy. Leave your email and we'll send you what moved for cohere-training-data-crawler: who started blocking, who stopped, and on which day.

Email me changes for cohere-training-data-crawler →

Work email only. No newsletter, no obligation.

Prefer to just buy it? See the sealed-archive monitoring offers, priced, with a dated sample.

What moved: cohere-training-data-crawler verdict changes

Counted domains whose robots.txt changed what cohere-training-data-crawler may do, between one sealed date and the previous one. Aggregate only — we do not publish which site.

Across the whole archive: 218 cohere-training-data-crawler verdict transitions, of which 130 moved TO a full block and 55 moved AWAY from one.

Sealed date	Verdict moved	Domains
2026-08-07	not mentioned → blocked outright	2
2026-08-07	not mentioned → partly restricted	1
2026-08-07	partly restricted → not mentioned	1
2026-08-07	blocked outright → not mentioned	1
2026-08-06	partly restricted → not mentioned	1
2026-08-06	blocked outright → not mentioned	1
2026-08-05	blocked outright → not mentioned	2
2026-08-05	not mentioned → partly restricted	1
2026-08-04	not mentioned → blocked outright	3
2026-08-03	blocked outright → not mentioned	1
2026-08-02	not mentioned → blocked outright	1
2026-08-01	not mentioned → blocked outright	2
2026-08-01	blocked outright → not mentioned	1
2026-07-31	blocked outright → not mentioned	19
2026-07-31	not mentioned → blocked outright	2
2026-07-30	not mentioned → blocked outright	3
2026-07-30	blocked outright → not mentioned	1
2026-07-29	blocked outright → not mentioned	1
2026-07-28	not mentioned → blocked outright	3
2026-07-28	blocked outright → not mentioned	1
2026-07-27	not mentioned → blocked outright	1
2026-07-25	not mentioned → partly restricted	1
2026-07-25	not mentioned → blocked outright	1
2026-07-25	blocked outright → not mentioned	1
2026-07-24	not mentioned → blocked outright	6
2026-07-23	not mentioned → partly restricted	2
2026-07-23	not mentioned → blocked outright	1
2026-07-23	blocked outright → not mentioned	1
2026-07-23	named and allowed → not mentioned	1
2026-07-22	not mentioned → blocked outright	8
2026-07-22	partly restricted → not mentioned	1
2026-07-21	blocked outright → not mentioned	3
2026-07-21	not mentioned → partly restricted	1
2026-07-21	not mentioned → blocked outright	1
2026-07-21	not mentioned → named and allowed	1
2026-07-20	not mentioned → blocked outright	1
2026-07-19	not mentioned → blocked outright	1
2026-07-18	blocked outright → not mentioned	1
2026-07-17	not mentioned → blocked outright	4
2026-07-17	not mentioned → partly restricted	1

Read this before quoting a trend. We hold 55 sealed daily snapshots between 2026-06-09 and 2026-08-07, out of 60 calendar days. 5 calendar day(s) inside that span have no collection run at all: 2026-06-11, 2026-06-12, 2026-06-19, 2026-07-04, 2026-07-15. Consecutive rows in every table below are consecutive SEALED dates, not consecutive days — where a gap falls, that row's diff covers more than 24 hours, and the span is printed with it. We do not interpolate across a day we did not collect.

Every sealed date

The denominator column is the number of panel domains that served a parseable robots.txt that day. It moves, so a raw count moves with it — compare the ratio, not the count.

Sealed date	Names it	Blocks outright	Partly restricts	Names & allows	Parsed that day
2026-08-07	1,046	910	108	28	61,552
2026-08-06	1,042	906	108	28	61,421
2026-08-05	1,043	906	109	28	61,527
2026-08-04	1,046	910	108	28	61,620
2026-08-03	1,045	909	108	28	61,533
2026-08-02	1,043	907	108	28	61,559
2026-08-01	1,042	907	107	28	61,326
2026-07-31	1,054	920	106	28	61,595
2026-07-30	1,068	933	107	28	61,444
2026-07-29	1,055	920	107	28	61,492
2026-07-28	1,063	928	107	28	61,562
2026-07-27	1,061	926	107	28	61,440
2026-07-26	1,063	928	107	28	61,573
2026-07-25	1,064	929	107	28	61,584
2026-07-24	1,061	927	106	28	61,576
2026-07-23	1,058	923	107	28	61,758
2026-07-22	1,060	926	105	29	61,744
2026-07-21	1,053	918	106	29	61,565
2026-07-20	1,051	917	106	28	61,665
2026-07-19	1,053	920	105	28	61,451
2026-07-18	1,053	920	105	28	61,782
2026-07-17	1,053	920	105	28	61,659
2026-07-16	1,051	919	104	28	61,526
2026-07-14	1,017	886	104	27	61,542
2026-07-13	1,014	882	104	28	61,448
2026-07-12	1,011	881	103	27	61,518
2026-07-11	1,011	880	104	27	61,526
2026-07-10	1,010	880	103	27	61,425
2026-07-09	1,008	880	101	27	61,495
2026-07-08	1,007	877	102	28	61,364
2026-07-07	1,006	876	102	28	61,491
2026-07-06	999	870	102	27	61,455
2026-07-05	1,001	871	102	28	61,562
2026-07-03	1,002	872	102	28	61,535
2026-07-02	1,005	876	101	28	61,553
2026-07-01	1,001	872	102	27	61,599
2026-06-30	1,001	872	103	26	61,590
2026-06-29	999	868	105	26	61,814
2026-06-28	994	863	105	26	61,835
2026-06-27	989	862	101	26	61,848
2026-06-26	993	864	103	26	61,851
2026-06-25	987	858	104	25	61,867
2026-06-24	990	862	103	25	61,660
2026-06-23	998	869	103	26	61,877
2026-06-22	991	863	102	26	61,911
2026-06-21	988	859	103	26	61,905
2026-06-20	987	858	103	26	61,754
2026-06-18	988	859	103	26	61,805
2026-06-17	992	865	101	26	61,952
2026-06-16	990	864	100	26	61,645
2026-06-15	991	865	100	26	61,715
2026-06-14	992	866	101	25	61,697
2026-06-13	989	864	100	25	61,648
2026-06-10	980	856	99	25	61,489
2026-06-09	973	849	99	25	61,599

Why you will not find a robots.txt file here. These pages render mechanical verdicts and aggregate diffs, never a fetched body. Site operators leave contact addresses in robots.txt comments — on the 2026-08-06 snapshot 2,200 of the distinct bodies we fetched contain an email-shaped string — and republishing bodies would republish those. Nothing on this surface identifies an individual site, either: every figure is a count.

Compiled from each site's own published /robots.txt, /ai.txt, /llms.txt and /.well-known/tdmrep.json, fetched daily over a frozen 100,000-domain research panel (source list: Tranco; no Tranco rank is republished here). Verdicts are mechanical readings of what those files say under RFC 9309 grouping with exact product-token matching — they are not legal advice, not a statement about any site's intentions, and not a claim about what any crawler actually did. A robots.txt is a request, not an access control. Figures are provided for informational purposes only and carry no warranty of accuracy or completeness.

AI crawler policy home · What changed · How this is collected · US Tech Automations.