What Would the Bill Require?
Its sponsor describes one core duty. Operators of automated software that “systematically access and copy content from UK websites” would have to “identify themselves and disclose the purpose for which they are collecting data,” according to a statement Damian Hinds released on June 29, 2026 (Hinds statement). Parliament’s bill page adds that the bill provides for the registration of operators of specified automated software and for duties on how that software accesses material published by others (UK Parliament). Parliament records its first reading on June 17, 2026.
Hinds is explicit about what it would not do. “The Bill does not regulate AI models, determine how automated software may use data, or change copyright law,” his statement says, adding that many major search engines and AI companies already identify their bots and that the bill is “primarily intended to address operators who currently do not.” In Press Gazette’s June report, Hinds put it this way: “if you run an online bot that accesses a website and takes content and data, you have to say who you are and what you’ll do with what you take” (Press Gazette, June 24). The same report says Hinds worked with the News Media Association to draft the bill. Parliament’s site would not serve the bill text to our requests, so we describe its mechanisms only from the long title, the sponsor and the trade press. For the controls a site has today, see how to block or allow AI crawlers in robots.txt.
What Did the Newspaper Bosses Ask For?
Press Gazette lists the signatories of the October 16 letter as Anna Bateson (Guardian Media Group), Chris Evans (The Daily Telegraph), Chris Longcroft (The Times and The Sunday Times), Danny Groom (Daily Mail), Dominic Carter (The Sun), Geoff Maynard (Daily and Sunday Express), Jon Slade (Financial Times) and Neil McIntosh (Daily Record), joined by executives from Condé Nast, the Liverpool Echo, Reddit and Mumsnet (Press Gazette). They urged the Prime Minister to back the bill, which Press Gazette says would have a better chance if “backed by Prime Minister Andy Burnham, who could also introduce a similar bill of his own.”
As Press Gazette summarizes the measures the letter describes, operators would state their identity, contact details and purpose; operators “could register to be on a ‘white list’ of good bots”; “online safety regulator Ofcom could publicly name and shame offenders by publishing compliance notices”; site owners could seek an injunction or damages; and “courts could order UK internet providers to block persistent offenders.” The letter argues that robots.txt, “the only established means for a website to signal which bots may access it,” “carries no legal force and is being ignored on a mass scale.” We did not see the letter itself, so these are Press Gazette’s quotations and summary.
The signatories have a commercial stake. That does not make their claims wrong, but the NMA’s figure that 19 of the 20 biggest UK news sites can be scraped by stealth tools, and the view that voluntary standards have failed, are positions in a lobbying effort rather than independent findings. For how the robots.txt picture looks from the outside, see whether news sites block AI crawlers.
What Happened on October 16?
The bill was not debated. Friday sittings are when Private Members’ Bills are taken, and on October 16 the Commons spent its debate time on another bill. The Automated Online Software (Access and Transparency) Bill appears in Hansard under “Business without Debate”: the motion “That the Bill be now read a Second time” was made, an MP said “Object,” and the record reads “Bill to be read a Second time on Friday 27 November” (Hansard). Hansard does not name the objecting member in that entry, and more than a dozen other bills met the same objection that afternoon. Parliament’s bill page still shows the bill at second reading (UK Parliament).
In practice, a single objection at that stage stops a Private Members’ Bill for the day without a vote. The bill survives, though it has lost a sitting. Press Gazette said in June that the bill is “unlikely to become law, unless it secures backing from the Government,” while noting it “could well help shape other legislation” (Press Gazette, June 24). Nothing we read shows the government has said whether it will back it.
How Much Bot Traffic Is the Bill Aimed At?
The most recent figures Press Gazette has run come from 51Degrees, which says it analyzed about three billion website visits in the year to May 29, 2026, on sites it works with in telecoms, advertising, media, retail, financial services and technology. It found AI bots made up 13% of web sessions in May 2026, up from 8% a year earlier, and that AmazonBot and Meta-External-Agent together accounted for 63% of sessions from the 72 AI-related bots it identified (Press Gazette, Sept. 3).
% of web sessions or visits on sites 51Degrees works with (as reported)
The total is for May 2026; the four bot rows cover the year to May 29, 2026, and Press Gazette describes AmazonBot’s share as “of all web visits” and the others as “of total web sessions.” “More than 4%” and “almost 3%” are charted at 4 and 3. The research covered AI-related bots only and excluded Google’s main search crawler. We did not read 51Degrees’ own release.
The chart shows that a few named, self-identifying bots account for most of the AI-bot traffic 51Degrees could attribute; the bill is aimed at the operators who do not identify themselves, who by definition do not appear in a list of named bots.
The same report cites Known Agents data that AI scraper and data providers abide by robots.txt requests 95.6% of the time and that the main gap is that publishers list only about 21% of AI scraper bots. That cuts in two directions: compliant bots are already controllable, and the unknown share is what neither dataset measures. For checking whether a bot is what it says, see how to verify a crawler is real; for how sitemaps fit in, see whether AI crawlers read your sitemap.
How Does This Compare With the US Efforts?
In Washington, Reps. Laurel Lee, Valerie Foushee and Gus Bilirakis introduced the bipartisan Stealth Bot Prohibition Act, H.R. 9915, which the News/Media Alliance says “would require AI stealth crawlers (‘bad bots’) to disclose their identity and purpose” (News/Media Alliance). In New York, Press Gazette reported in June that the Stealth Crawler Prohibition Act had passed the Senate and Assembly and “now only needs to be signed by the state governor,” and would make it an offence to “damage, impair or burden the operation of a covered news site” (Press Gazette, June 24). We could not confirm whether it has since been signed, so treat its status as unknown. The UK bill, as its sponsor describes it, is the narrowest of the three: identification and registration rather than new offences against scraping itself.
Publishers are pursuing other routes at the same time. Our earlier pieces cover Reach’s move against OpenAI’s crawlers, USA Today Co.’s lawsuit against OpenAI and Meta’s new crawling clause for business accounts.
What This Means for Anyone Buying a Placement
These are our recommendations, not findings of the sources. First, nothing has changed yet: no UK law requires bots to identify themselves, and the bill’s next chance is November 27. Second, the outlets where companies seek coverage are the ones pressing for bot rules, and their licensing and blocking choices affect whether their articles reach AI tools at all; a placement in a title that blocks AI crawlers may be less visible in AI answers, and one in a title that licenses content may be more so, but no source we read measures that for a given article. Third, a bot register would tell a publisher who is copying its pages; it would not tell a company whether an AI answer cites its coverage. That is a separate question, covered in which sources AI assistants cite. Fourth, no placement, paid or earned, guarantees inclusion in or exclusion from AI answers, and a paid article remains advertising that must be labeled.
Methodology and Limitations
Four limits apply. First, we relied on Press Gazette for the letter, its signatories and its wording, and did not see the letter. Second, Parliament’s publications site would not serve the bill text to our requests, so the mechanisms come from the bill’s long title, the sponsor’s statements and Press Gazette; which enforcement powers are in the bill and which are the letter’s requests is not settled here. Third, the second-reading outcome comes from Hansard’s record of business without debate, which does not name the objecting MP. Fourth, the traffic figures are 51Degrees and Known Agents data as reported by Press Gazette, from samples of sites those firms work with, not the whole web.
Conclusion
The UK proposal asks for something narrow, that bots say who they are and why they are copying, and it has the national press behind it. It also has the usual problem of a Private Members’ Bill: one objection on a Friday was enough to push it back six weeks. For a company, the practical point is that the rules on AI scraping of the outlets it pitches are being argued, not settled.
Frequently Asked Questions
What is the Automated Online Software (Access and Transparency) Bill?
It is a UK Private Members' Bill sponsored by Conservative MP Damian Hinds that would require operators of bots that systematically copy content from UK websites to identify themselves and say why they collect it. Parliament gives its long title as a bill about the registration of such operators and their duties.
What happened at the bill's second reading on October 16, 2026?
It was not debated. Hansard records an objection when the second-reading motion was made, and the bill was set down for a second reading on Friday, November 27, 2026.
Who is the UK Prime Minister the publishers wrote to?
Andy Burnham, who GOV.UK says became Prime Minister on July 20, 2026. Press Gazette reported the letter from eight national newspaper bosses on October 16.
Would the bill ban AI scraping?
No, according to its sponsor. Damian Hinds says it does not regulate AI models, determine how automated software may use data or change copyright law; it sets transparency requirements.
Does robots.txt stop AI bots?
It is a voluntary signal with no legal force, as the publishers' letter says. Known Agents data cited by Press Gazette says AI scrapers abide by robots.txt 95.6% of the time but publishers list only about 21% of AI scraper bots.
Is there a similar law in the US?
A federal Stealth Bot Prohibition Act, H.R. 9915, has been introduced, and New York's legislature passed a Stealth Crawler Prohibition Act that needed the governor's signature as of June. This is not legal advice.
Sources
This guide is built from Parliament's bill record and Hansard, the sponsor's statement, GOV.UK, three Press Gazette reports and a News/Media Alliance release, all read on October 18, 2026. Advice for buyers is our recommendation. We did not read the bill text, the publishers' letter or 51Degrees' own release.
- official UK Parliament, Automated Online Software (Access and Transparency) Bill, bill record (via the Bills API). Supports the long title, sponsor, first reading on June 17, 2026 and current stage. Read October 18, 2026.
- official Hansard, House of Commons, Business without Debate, October 16, 2026. Supports the objection and the new second-reading date of November 27, 2026. Read October 18, 2026.
- official Damian Hinds MP, Damian's proposed law to support publishers and journalism, June 29, 2026. Supports the bill's purpose and scope, the quoted passages and the scheduled October 16 second reading. Read October 18, 2026.
- official GOV.UK, The Rt Hon Andy Burnham MP. Supports that Andy Burnham became Prime Minister on July 20, 2026. Read October 18, 2026.
- trade press Press Gazette (Charlotte Tobitt), Guardian and Sun agree on ban for deceptive AI bots, October 16, 2026. Supports the letter, signatories, quoted wording, requested measures and the NMA 19-of-20 claim. Read October 18, 2026.
- trade press Press Gazette (Charlotte Tobitt), UK news industry backs law to stop deceptive AI scraping, June 24, 2026. Supports the bill's prospects, NMA drafting role, Hinds quote and the New York bill details. Read October 18, 2026.
- trade press Press Gazette (Alice Brooker), Amazon and Meta AI bots account for most website visits, September 3, 2026. Supports the 51Degrees and Known Agents figures and the chart. Read October 18, 2026.
- trade body News/Media Alliance, News/Media Alliance Applauds AI Bot Transparency Bill Introduction, July 23, 2026. Supports the H.R. 9915 sponsors and description. Read October 18, 2026.