What has shipped

Chat checks report access blocks and unconfirmed widget loading

When the server gives the scan an access block or a browser challenge, the chat check now reads Unverified and gives that reason. Chat code without a confirmed working widget also reads Unverified, where the report used to say no widget was detected. A working widget that reports itself hidden stays N/A. Talkdesk widgets are now recognised, including buttons that load their chat code only when clicked. Language coverage considers alternate links on the scanned host rather than other country sites. Reports graded under earlier packs keep their recorded results.

A chat widget is credited to LiveChat only when LiveChat's own code is on the page

The scanner credited a widget to LiveChat whenever the page had an element with the id chat-widget-container. Other chat products use that id too, and some pages carry the container with nothing in it, so sites with no LiveChat code were reported as running LiveChat, and an empty container was reported as a chat widget the scanner could not open. From the page markup alone, a page now counts as LiveChat only when it carries one of the two frames LiveChat itself draws. A site's own chat inside such a container is handled like any other widget we do not recognise, and an empty container no longer counts as a widget. Real LiveChat installs read as before. Grading is unchanged.

A typing indicator no longer ends the wait for a chat's first message

The wait added on 6 October stopped once the panel's text held still, and a "Typing" line holds still while its dots animate, so a scan could end before the assistant's greeting arrived. A typing indicator now means the message is still being written. A second correction concerns widgets whose panel matches more than one element on the page. The text was read from the first match, which on some Userlike installs is a frame one pixel high sitting ahead of the chat, so an open panel read as empty and the chat check went Unverified. The text now comes from the panel the scanner found open, the one its screenshot shows. Grading is unchanged.

Chat checks wait for the first message, and say when a consent wall hid the page

The scanner read a chat panel the moment it appeared. Some widgets draw their greeting a beat later, so a scan could record the panel's header and input box without the message a visitor reads first. It now waits until the panel's text has stopped changing, for at least two seconds and at most five, and a panel still showing "loading" at that point counts as unreadable. Consent walls that ask a visitor to consent or subscribe are now accepted as well, including Sourcepoint walls served from the publisher's own domain. When a wall still covers the middle of the page after that, the chat check reads Unverified and gives that reason, where it used to report that no chat widget was found on a page the scanner never saw. That clause is the only grading change from 2026.10, and reports graded under earlier packs keep their results. Two smaller corrections shipped with it: a report with checks left for manual verification now leads with them, ahead of any "No gaps detected", and a page at /ai counts as an AI policy page only when its own title says it is one, because that path usually holds an AI news section or a product page.

A text-label check with nothing to act on now reads N/A instead of Unverified

On an article page with no AI label, where nothing on the site signals AI-generated text, the text-label check came back Unverified, although it had run and found nothing to act on. It now reads N/A with its own explanation: we cannot tell from outside whether text was written by a person, we do not guess, and AI-generated text needs a label. Every other grading rule is unchanged from 2026.09, and reports graded under earlier packs keep their original results. A report where every check is N/A now says that nothing on the pages scanned triggered an Article 50 check, and monitoring stops alerting when a row changes only because the rules did.

An article page carrying the EU's AI icon no longer reads "no AI label" in the coverage table

The page-coverage table of a crawl printed “no AI label” for an article page whose only label was the EU AI icon, while the grade beside it already counted the page as labeled. The row now reads “labeled (EU icon, filename match)” or “labeled (EU icon, alt-text match)”. On inner pages the icon is recognised by its file name or alt text; the scanner confirms the icon's exact file only on the homepage. Grading is unchanged.

An unreadable chat widget now comes with the reason, and more widgets open

When the three chat checks could not read a widget, the report said so without saying why. It now prints what the scanner saw: its chat button was found and clicked and no chat panel appeared, or another element was lying on top of the button. Every grading rule is byte-identical to 2026.08.2; only those three report texts changed, so a report regraded across the two versions keeps its states. The same day, the way widgets are opened changed. Every element that matches a chat panel is now tested instead of only the first, candidate buttons are clicked smallest first, Zendesk panels are recognised in their translated names, and the vendor recipes went from 20 to 22 with moinAI and Trengo.

A Hungarian term the review killed had shipped anyway, and is now out

When the disclosure lexicon was extended to 24 languages in July, every proposed term went through a cross-language collision review, and one Hungarian entry was recorded as killed: hyphens in our matcher also match spaces, so the term for an AI assistant, MI-asszisztens, really matched mi asszisztens — and mi is the everyday Hungarian word for "we", while asszisztens is a human job title. A clinic writing "a mi asszisztens csapatunk" ("our assistant team") about its own staff would have been graded as disclosing an AI. The kill was recorded; the term shipped anyway. We found the slip this week while preparing the lexicon for open-source release — the pre-publication review checked the shipped set against the review record, and they disagreed. The term is now removed, with the sentence above pinned as a regression test. Language coverage is unchanged: Hungarian keeps its verified terms, and the idiomatic greetings this entry was meant to catch nearly always carry mesterséges intelligencia or chatbot anyway.

The State of AI Disclosure 2026, our first published study

Eight days after Article 50 began to apply, we swept a locked cohort of 1,142 detector-flagged, EU-facing sites and published the aggregates: how many of the most-visited EU sites show a real chat launcher once our detector's measured error is corrected out, how often an automated visitor can read a widget's first message at all, and what share of the readable first messages disclose the AI. Every number travels with its methodology — the exclusion funnel, the pre-registered timestamp anchors, what went wrong on the way (a retired baseline, a noise floor never obtained, both stated rather than smoothed over), and our own instrument's measured errors, including a blind re-audit of our largest failure bucket that found two-thirds of it was our detector admitting things that were never chat widgets. Aggregates only: no site is named, and none will be. Read the study →

Irish joins the disclosure lexicon, and the scanner now says when it cannot read the language

The disclosure lexicon covered 23 of the 24 official EU languages, plus Catalan; Irish was the gap. Its terms were verified against the national terminology database and real Gaeilge usage rather than machine-translated — intleacht shaorga, bota comhrá, and their declined forms — with the same cross-language collision review every other language went through (the Irish abbreviation for artificial intelligence can never be matched: it is spelled exactly like the English word "is"). Irish cookie-consent buttons are now dismissed too, so Irish-language sites' widgets get opened and read at all. And a new outcome closes an old gap: when a page's own markup declares only languages the lexicon does not cover, an unmatched greeting is now reported as could not assess (language) instead of "no disclosure detected" — the latter was a claim about text the scanner cannot actually read, and our own study review called it out.

The chatbot probe stopped mistaking your own message for a reply

When the scanner types a question into a live chat widget to see whether it answers like a machine, it now waits for the widget to echo that message back before it starts timing anything. Nearly every chat widget renders what you just typed within a fraction of a second, and the probe had been reading that echo as the assistant's reply, which made almost any widget it could open look automated. Confirmed automation is the one signal that can turn a hedged finding into a plain fail, so the cost of getting it wrong is a fail on a site that never earned one. It is now measured from the echo forward, and a candidate reply that merely repeats our own words is discarded.

Three other reads were tightened for the same reason. A page stating that its content is not AI-generated used to be counted as carrying an AI label, because the phrase matched either way. Article 50(4) now gained a clause for the case where a publisher declares AI involvement themselves: that used to fall through to a report line saying no signal was found, which was untrue on the page it was printed on. Image metadata reading the Spanish word for image, or the Italian for "from the", was being matched against the names of AI image tools. And a chat greeting saying "I am not a human, I am an AI assistant" was being read as no disclosure at all, because the denial of being human was taken to cancel the disclosure that followed it. French, Spanish and Italian gained the compound phrasings those languages actually use.

Article 50(4) reports its evidence chain

Reports, and the Evidence Pack PDF, now show the evidence chain behind an AI-content finding: which signals the site itself emitted, what kind each one is, and where we found it. A generator tag, provenance on published images, and an AI policy page describing how content is produced are three independent kinds of evidence; the same tag repeated across five hundred pages is one. Strength is measured that way, by independent kinds rather than by count, so a single template cannot inflate a site into looking heavily flagged. Every row is something you can go and check yourself. We still do not run automated AI-text detection and still publish no likelihood score: a percentage would read as a measurement of the writing, and no such measurement is reliable enough to seal into evidence.

Article 50(4) now reaches a clear fail, and stops reporting what it cannot see

This check now has three distinct answers instead of one. When a page's own markup declares its content machine-generated and the reader is shown no disclosure, that is a fail: the publisher has stated it, we simply compared that against what a visitor sees. When a site declares a machine writer site-wide but nothing ties it to a given page, that is flagged for review, because a generator tag does not establish which pages the tool produced. And when there is no machine-readable signal at all, the check says plainly that it could not determine who wrote the text, instead of reporting every unlabelled article as a finding. On an ordinary site with no AI involvement the old behaviour flagged nearly every page, which is noise dressed up as a finding. We do not run statistical AI-text detection, and we do not intend to: it is not accurate enough to seal into evidence, and a wrong answer would be an accusation carrying our signature. The label vocabulary also grew to cover the natural phrasings publishers actually write, in all six languages, including text using typographic apostrophes.

Crawls read your whole sitemap, and respect more of your robots.txt

Large sitemaps are no longer read only in part, and a crawl now records whether it saw all of your sitemap so a coverage claim can never overstate itself. Robots directives using wildcards or query patterns, such as disallowing tracking parameters, are now honoured rather than ignored. Page ordering was rebuilt too: contact, pricing, about and policy pages are picked first, and article pages follow, instead of a headline that happened to contain the word "pricing" outranking your actual pricing page.

Anyone can now check an evidence pack

A new verification page takes an evidence manifest and answers two questions separately: whether the evidence is unaltered, which you can prove yourself with a plain SHA-256 and no help from us, and whether we sealed it, which is a shared-secret signature only we can confirm. The page says plainly which of the two relies on trusting us. Verification keeps working after a scan has aged out of retention, because the signature check is pure arithmetic over the bytes you hold.

Reports are organised by duty

Every report now opens with a summary table: one row per Article 50 duty, the checks it covers, and where they landed. The detail below is grouped the same way instead of running as one flat list, so you can answer a question about synthetic media without reading the chatbot findings first. Printing a report from the browser now produces a usable document.

A real dashboard on every plan

The dashboard gained a sidebar and several new sections for all accounts, free included: open findings gathered across every site you own, an Article 50 readiness view explaining each duty and where your latest scans stand against it, and a disclosure toolkit with wording in the six languages the scanner reads, checked against the scanner's own lexicon so the text it gives you is text it recognises.

Better reading of cookie walls

Consent banners are the main thing standing between a scan and a chat widget's first message. The scanner now recognises fourteen more consent platforms, reads accept buttons in ten languages rather than English only, and can see banners that render inside shadow DOM. It still refuses to click anything that only accepts necessary cookies, since dismissing a wall without granting consent would hide the widget just as effectively.

Official EU icons and structural C2PA

Added detection for the European Commission's official AI-content icons (the Annex I set from the 10 June 2026 Code of Practice) and structural inspection of C2PA Content Credentials, not just a mention of them. The guide library grew to twelve chat-widget vendors and eleven EU member states, each researched to its enacted-versus-proposed status.

Multi-page coverage

Scans can look past the homepage to the pages where Article 50 duties tend to live: contact, pricing, article and policy paths first. The report says exactly which URLs were checked, so the coverage claim in an evidence pack is honest.

Owner verification

Prove you control a domain with a DNS TXT record or a meta tag. A verified owner can scan their own site more deeply and keep evidence longer.

Sealed, tamper-evident evidence

Every scan now hashes its findings and captured artifacts into a manifest and signs it (HMAC-SHA256). A report can be re-verified later, and any change to the underlying evidence breaks the seal.

The six core Article 50 checks

Chatbot disclosure at first interaction (matched against a six-language lexicon) and its persistence on mobile, machine-readable media marking, AI-content labels on article-like pages, an AI-policy-page check, and the emotion-recognition and biometric-categorisation notice all went live. Each is a deterministic check graded against the versioned pack and reported as detected, not detected, or needs verification.

DisclosureProof launched

The free homepage scan, the plain-language Article 50 guide, and the per-vendor and per-country reference library went live.