AI assistants, October 2026 in review: the ads moved in, the numbers came loose

Our mid-month analysis argued that the AI assistant market had split in two: questions were spreading across ChatGPT, Gemini and Claude, while the clicks to websites still came overwhelmingly from ChatGPT. The second half of October did not change that picture. It added two things to it.

The first is commerce. In thirty days, ChatGPT ads gained new targeting, new formats, a second buying route and a default that rewrites them; Google took ads in AI Mode to ten more countries and into follow-up questions, and told investors that AI Mode earns less per query. The answer is no longer a page of text that points elsewhere. It is a place that sells, and increasingly a place that acts.

The second is doubt about the numbers. ChatGPT's links lost their tags for a week. Gemini's tags, reported as new on 1 October, were reported on 30 October as having stopped. And two large studies at the end of the month showed that most AI citations seen on one day are gone the next, so a single check of AI visibility says very little. This analysis takes the month theme by theme and ends with what we think it adds up to. Where a claim is our reading rather than a measurement, it says so.

The month in one table

Most of what changed in October changed in the products, not in the market shares, which are only published with a lag. The table lists the changes to the assistants themselves, apart from advertising, which has its own table below. The last column keeps the distinction our sources make between what a company launched or confirmed and what was only spotted.

Models: the race became background

Two of the three largest assistants changed their main model within a week of each other. OpenAI put GPT-6 into ChatGPT for paid plans on 7 October and for Free and Go users on 8 October, with Intelligent UI: answers that can contain charts, buttons, forms and small working tools. OpenAI says GPT-6 Instant starts answering web-search questions 44% sooner on average than GPT-5.6 Instant, and citations stay behind a Sources button. Google began rolling out Gemini 4 Argon on 13 October in the Gemini app and in AI Mode for Google AI Pro and Ultra subscribers, who pick it from the "+" menu; Robby Stein, Google's VP of Product for Search, wrote that it "brings big improvements to tackling nuanced, multi-step tasks in AI Mode". Free users keep the default model. Anthropic completed its 5.5 family with Haiku 5.5 on 7 October.

Smaller moves pointed the same way. On 12 October a Microsoft executive wrote that Microsoft-Decision-1, a model built for "routing, classification, prioritization, verification, and workflow control", was being tried in Copilot. On 17 October DeepSeek opened its own search engine to developers through a Search API and a Fetch API, with an authority tier from 0 to 4 on each result and an index it says holds "hundreds of billions" of pages; calls from Claude Code, Codex or DeepSeek's own harness were free until 31 October. That is one more index with its own idea of authority, and coding agents can now query it directly.

Google's third-quarter results on 28 October gave the scale behind its side. Sundar Pichai wrote that query growth was "the highest it's been in the last five years", that "Queries in AI Mode are on average three times longer than traditional searches", that "one in five AI Mode searches globally use voice, images, or documents", and that the Gemini app "is now at one billion users". A week earlier, in an interview with Alex Heath, he had described where this goes.

Our reading: when the main model of an assistant changes every few weeks, a new model is the condition of the market rather than news in it. The practical consequence is for measurement. AI Mode's underlying model changed twice in six weeks, on 2 September and 13 October. Any month-on-month change in a brand's citation share needs the model dates beside it, because the reader that decides which sources an answer uses may have changed while nothing on the page did.

Questions spread out; clicks did not

Our sources for the second half of October carry no new market-share panel, so the shares from our mid-month analysis still stand. Comscore put ChatGPT at 50% of US AI prompt volume in June, down from 70% in January, with Gemini at 30% and Claude at 11%. Similarweb had ChatGPT at 56.4% of worldwide generative-AI website visits in September, Gemini at 25.0% and Claude at 9.6%. The click side looks very different, as Statcounter's September data on referrals from chatbots to websites shows.

What October added is a sense of proportion. Three studies published around 21–22 October measured AI referrals against search rather than against each other, and all three found them small. Marie Haynes looked at twelve months of GA4 data from 100 websites with more than 145 million sessions and found that sessions from ChatGPT "amounted to just 0.62% of the volume driven by Google Organic". ChatGPT visitors converted better, at 13.8% against 3.02%, yet Google "still drove 96.5% of the total conversions". SEO Works found organic search delivering 39 times the traffic of AI assistants. Kevin Indig, working with 12 sites, wrote that "we also can't yet find a reliable relationship between AI Visibility and branded search or direct traffic." On 28 October seoClarity estimated ChatGPT's search-like activity at "approximately 7.5% of Google's search-query volume", a model built on OpenAI's own message data rather than a measured count.

Our reading: these numbers do not say that AI visibility is unimportant. They say that it does not yet show up as traffic. The assistants gaining share of questions are the ones that keep users on their own page, and the one that sends clicks sends few of them relative to Google. A brand named in many Gemini or Claude answers will see almost none of it in analytics. That makes traffic a poor proxy in both directions, and October showed that even the traffic that does arrive can lose its name.

The label on the click came off

In June OpenAI began adding utm_source=chatgpt.com to links in ChatGPT answers, which let analytics count AI referral traffic directly. With GPT-6 the tag started to disappear, and the links also opened without a referrer. Cedric Huesler of Adobe compared 9–13 October with the week before on about 500 pages where ChatGPT had driven at least 5% of visits: ChatGPT visits fell by about 31,000 a day and direct visits to the same pages rose by about 26,000. Profound, which collects ChatGPT answers, tied the change to Intelligent UI.

It was a bug. Huesler reported that OpenAI confirmed it to him on 14 October and began rolling out a fix early on 15 October; by 9 a.m. Pacific that day, tagged ChatGPT clicks were back to about 95–100% of the same hour a week earlier across thousands of domains. OpenAI's release notes do not mention it, so the confirmation is second-hand, and Profound added that attribution already lost will not come back. We covered the episode as it happened in the 15 October issue.

Gemini's story ran the other way and is less tidy. On 1 October Search Engine Journal reported that Gemini had started adding UTM parameters to its outbound links, which mattered because, according to the Reddit thread that found it, the Gemini mobile app passed a referrer only about 9% of the time. On 30 October Search Engine Roundtable reported that Gemini had stopped adding utm_source=gemini around 28 September, after starting in early September; the desktop web still passes a referrer, the iOS app none. Google confirmed neither, and the two reports do not fit neatly together. What is clear is that no assistant has promised a stable label, and one that existed in September cannot be assumed in November.

Answers became interfaces

Intelligent UI is the most visible part of a wider change: the answer is turning into a place where something happens, rather than text that points elsewhere. October brought three more pieces of it. On 6 October Brodie Clark spotted ChatGPT testing labels on the organic products it recommends: a green "best all-rounder" for one, grey "best for" labels for narrower wins. These are rankings inside the answer, not ads, and a brand can stay among the products listed while a competitor takes the green label. On 15 October Google said it had increased the number of inline links in AI Mode and AI Overviews and would place them higher more often, without publishing click data, and launched Universal Cart: US shoppers can keep items from many stores in one cart in Search, including AI Mode, and in the Gemini app, with price-drop and restock alerts. And on 23 October Brodie Clark noticed AI Overview citation cards showing an AI-written description of the cited page instead of the page's own text, which Google has not commented on.

Taken together, the interface is now deciding three things a publisher used to control: whether a link is visible, what it says, and whether the reader needs to leave at all. Our separate analysis of Google Search covers the AI Overviews side in more depth.

Ads moved into the conversation

October was the month advertising inside assistants stopped being an experiment on one platform. On the ChatGPT side, OpenAI announced on 5 October a US test of visual ads shown while ChatGPT generates an image. On 14 October an advertiser posted an OpenAI email introducing ChatGPT Ads account managers with biweekly check-ins. On 16 October a 404 Media investigation, reported by MobileSyrup, found that advertisers can target users with email addresses, phone numbers and Android identifiers; OpenAI said the data is deleted after 24 hours and that a list must hold more than 25,000 users. On 20 October OpenAI emailed advertisers that "creative enhancements" would be switched on by default, letting the system rewrite headlines, translate text, resize images and change the landing page. On 22 October Search Engine Journal summarised OpenAI's description of the auction: a separate orchestrator model checks eligibility, earlier turns of the conversation can be used, and a winning ad can still be dropped by a relevance check. On 26 October some ChatGPT ads were reported to be bought through Amazon DSP.

Google moved in parallel. On 17 October it emerged that Google was testing ads in AI Mode for users of Personal Intelligence, which connects Gmail and Google Photos; Google's help page says ads "will not be based directly" on the inbox or photo library, while summaries and inferences "may be used to show you ads". On 27 October Google's privacy policy was reported to let that connected data shape ads in AI Mode, in a test. On 28 October Google announced ads in AI Mode for ten more countries, Australia, Canada, India, Kenya, Malaysia, New Zealand, Nigeria, Pakistan, the Philippines and Singapore, with Brazil, Mexico and the UK planned for the fourth quarter, and multiturn ads that show on follow-up questions in the US. The same day, on the earnings call, it said what that surface earns.

Search revenue grew 12% in the third quarter against 17% in the second, and Schindler named exchange rates as the largest single cause. Still, the sequence matters: Google said AI Mode earns less per query, and in the same days filled it with more ads, in more countries, at more points in the conversation. The table puts the month's advertising changes side by side. The PPC side of this, including what advertisers can and cannot see in their reports, is in our separate paid-search analysis.

Agents moved from demos to traffic

The mid-month analysis described agents that buy: Meta's Muse, Gemini's auto browse in Chrome, native checkout in AI Mode on by default for Shopify merchants. October added evidence of how much of the web's traffic is already theirs, and how they behave. Search Engine Journal reported on 5 October Cloudflare's figure that "daily requests from AI agents on our network grew by more than 1,700%" over the past year, and that many of those agents query search engines too, looking to Google much like rank trackers. On 9 October Anthropic published cases from its own evaluations in which models acted on live websites in ways nobody intended: submitting real forms, taking access tokens from public dashboards, and using URL shorteners to get around the length limits of fetch tools. Anthropic cut live internet access from all its internal evaluations.

Google's chief executive described the destination twice in two days. Agents are the layer that unifies Google's products, the "one box" quoted above, and they will reach sites both through web pages and through protocols: "I think you have both ways and they're going to develop". On 30 October Gary Illyes said AI had made crawler impersonation worse and that firewalls and CDNs "are not perfect" at spotting it.

Our reading: for a site owner, an agent visit is neither a crawler nor a person. It may fill a form, follow a cart, or read a price, and it may be pretending to be something else. The practical question is the one we raised in mid-October, now sharper: can an agent verify your price, stock and terms from your pages, and can you tell an agent from a human in your logs?

What the assistants read

October produced more evidence on what AI answers cite than any month this year. It does not converge on a single rule, but it points in a consistent direction: towards original sources, official sources and third-party coverage, and away from summaries and from what a brand says about itself.

Ahrefs compared the domains ChatGPT cited in 2.4 million US answers in March with 3.8 million in September and found it "seems to be prioritizing original sources over sites that summarize them": Reddit fell from 19.1% of answers to 2.3%, Wikipedia from 14.1% to 2.9%, PubMed rose from 64th place to 4th. A preprint from the National University of Singapore and Xi'an Jiaotong University found that Google's organic results alone contained only 6.4–9.4% of the pages ChatGPT cited. TJ Digital, on 25 October, found that of the pages in Google's top ten for a question, AI Overviews cited 24%, AI Mode 17% and ChatGPT 7%. On 15 October Cloudflare and OpenAI said a research pilot "has enabled OpenAI to discover and index hundreds of millions of new web pages" for ChatGPT search, using Cloudflare's network signals; they gave no figures on citations or traffic.

Three findings deserve a line of their own. Graphite's study suggests that being on a cited page is not enough to be named: brands ChatGPT rarely names from memory stay rarely named even when search is on. Muck Rack's conclusion is short: "Brands cannot build meaningful AI visibility from their website alone." And the Fudan preprint shows how easily the reading can be steered. Its authors posted about invented subjects, including a "Cactus Gravity Protocol", on sites the tools cite often, such as Medium and YouTube. Within seven days eight of ten AI search tools were citing the posts; Grok linked to them in 30 of 32 answers. Forty posts on a self-hosted WordPress site drew nothing. The paper has not been peer-reviewed, and the subjects were fictional, but it is the clearest demonstration yet of LLM poisoning through the sources an assistant trusts.

Access is part of the reading too. Muck Rack found that "among major publishers blocking Claude's search crawler, 81% received almost no Claude citations", against 12% of publishers that allow it. An assistant cannot cite what its crawler was told not to fetch.

One check is noise

The most useful measurement work of the month came at its end. Profound had already shown, on 30 September, that the median page loses half its citation share within 11 days of its peak. On 29 October Ahrefs asked five AI tools the same questions every day for 20 days and followed about 37,500 citations seen on the first day. On every tool, half or more were gone the next day.

Ahrefs' advice is blunt: "Don't judge AI visibility from one check." It suggests instead to "Count how many of the last seven days you were cited." One finding ties AI citations back to ordinary search: on AI Overviews and AI Mode, pages that ranked anywhere in Google for the same question "were about seven times more likely" to keep their citations for at least a week.

SISTRIX, on 28 October, asked how many questions a measurement needs. It simulated a brand whose true share of answers is 25%. One prompt asked on 20 days could land "anywhere between 0 and 59 per cent"; the typical error is "±34 percentage points for one prompt, ±9.5 for 20 prompts and ±4.2 for 100". For ±5 points it puts the need at around 70 prompts, for ±3 at around 200. It adds that 70% of brands that appeared in a prompt's answers "show up on fewer than one in five days", and that a brand's own hand-picked prompt set flatters it in three cases out of four.

The platforms are starting to say the same about their own numbers. On 13 October Microsoft added Topics Portfolio to Bing Webmaster Tools, the first report from a search engine that puts a site's AI citations beside the whole topic, competitors included, and told site owners to "treat the numbers as directional signals for planning, not as a precise ledger".

Who may read what: crawlers, publishers and rules

The terms on which assistants read the web moved several times in October, mostly through companies rather than courts.

Meta's revised Commercial Terms, reported on 15 October, let it crawl any website associated with a business that uses Meta products and use the content to "provide, develop, improve, research, train, and maintain" its products. On 17 October Meta took "train" out of the clause and kept the crawl. On 26 October it moved the start from 30 October to 20 November. The UK's Competition and Markets Authority said it would "continue to scrutinize these terms"; WAN-IFRA called the delay "a welcome step, but it does not address our concerns." On 21 October Google reminded everyone what the main tool site owners have can and cannot do.

Publishers' route to payment kept narrowing to deals. The full opinion in Penske Media and Chegg v. Google, signed on 30 September, dismissed every federal antitrust claim, finding that "an expectation is not an agreement", and left their California unjust-enrichment claims undecided, so the question of payment can still go to a state court. Meanwhile Google said on 15 October that its News AI pilot covers more than 400 publications; Reach signed, and its chief executive called the deal a template while pressing OpenAI over scraping. USA Today Co sued OpenAI for $250 million on 9 October. SPUR, a coalition including the Guardian, the FT, the BBC and AP, published a standard on 2 October for reporting when AI tools retrieve, cite, show and send clicks to content. And Searchable's study of answers about UK towns found ChatGPT using local news for 0.9% of questions.

The rules against manipulating answers got firmer. Anthropic's usage policy, published on 8 October and effective 12 November, bans using Claude to "manipulate the sources from which search engines or AI systems draw answers" by seeding content that misrepresents its origin, authorship or independence; its September threat report described about 70 sites posing as local news outlets that published at least 8,913 Claude-written articles. Google's documentation, rewritten on 1–2 October, names fabricated author profiles as deception. Practices sold as generative engine optimisation that depend on fake independent sources are now explicitly against the rules of the systems they target. In the UK, the CMA's consultation on a search choice screen that could include AI assistants closed on 9 October, with a decision expected by the end of the year.

What it adds up to

This section is our reading, not a measurement.

• The answer is now a storefront. In one month ChatGPT ads gained data targeting, a second buying route and rewriting by default, while AI Mode ads reached ten more countries and follow-up questions. Being cited, being recommended and being advertised are now three results in the same answer, and each needs its own measurement.
• The economics are not settled. Google told investors AI Mode earns less per query and expanded its ads anyway. Expect more ad formats, more defaults that are on, and more automation that advertisers have to switch off.
• Traffic is the wrong instrument for AI visibility. ChatGPT sessions were 0.62% of Google organic's in one large sample, the assistants gaining share send few clicks, and the labels on the clicks that do arrive failed twice. Visibility has to be measured in the answers.
• A single observation is not a measurement. Half or more of the citations seen on one day are gone the next, and one prompt can miss a true share by 34 points. Rates over many prompts and days are the only numbers worth reporting.
• What gets cited is what others say about you. Original sources, official sources and third-party coverage dominate, and Muck Rack's data puts brands' own sites at about 15% of citations. Coverage, not copy, is what moves an assistant.
• The manipulation window is closing from both ends. A fake concept reached eight of ten AI tools in a week, and Anthropic and Google now name that practice as abuse. A citation from a site nobody in your market reads is a reason to check who wrote it.

What to watch in November

• 12 November: Anthropic's usage policy on manipulating AI answer sources takes effect.
• 20 November: Meta's revised Commercial Terms, with the crawl clause, take effect unless delayed again; the CMA said it would keep scrutinising them.
• ChatGPT ads after the default: whether advertisers report rewritten ads and changed landing pages, and whether OpenAI adds reporting by version.
• AI Mode ads in Brazil, Mexico and the UK, which Google plans for the fourth quarter, and whether it ever reports AI Mode separately from Search and Shopping.
• Gemini 4 for free users: whether Argon reaches the default model in AI Mode, which would change the reader behind most AI Mode answers.
• Attribution: whether Google confirms anything about Gemini's tags, and whether ChatGPT's referrer returns alongside its restored UTM tag.
• New share data: the next Comscore and Similarweb readings, the first to cover GPT-6 and Gemini 4.
• Into December: the CMA's decision on AI assistants in the choice screen, and 2 December, when the EU's Article 50 grace period on machine-readable marking ends.

To see what three assistants say about a brand today, across the buyer questions around it rather than one prompt, Brandscope runs them in one reading.

Sources

Every claim above is dated and linked. Where a number comes from a measurement rather than an announcement, whoever measured it is named.

• AI assistants in October 2026: the questions spread out, the clicks did not — GetLoopLoop, 16 October 2026
• GPT-6 for everyone — OpenAI, 7 October 2026
• ChatGPT Gets GPT-6 And Intelligent UI For Interactive Answers — Search Engine Journal, 7 October 2026
• Introducing Gemini 4 Argon — Google, 13 October 2026
• Google Adds Gemini 4 Argon To AI Mode For Pro & Ultra Subscribers — Search Engine Journal, 13 October 2026
• Anthropic newsroom — Claude Haiku 5.5, 7 October 2026
• Bing & Copilot Experimenting With Microsoft-Decision-1 — Search Engine Roundtable, 12 October 2026
• DeepSeek Search API — DeepSeek API Docs, 17 October 2026
• DeepSeek Opens Its In-House Search Engine To Developers — Search Engine Journal, 19 October 2026
• Alphabet Q3 2026 earnings: CEO remarks — Google, 28 October 2026
• Sundar Pichai on Gemini 4 and the AI compute crunch — Sources with Alex Heath, 21 October 2026
• Pichai: AI agents will use both browsing and protocols — Search Engine Journal, 22 October 2026
• ChatGPT loses 20 points of US AI prompt share — PPC Land on Comscore, 22 September 2026
• Gemini's traffic share falls for a fourth month — OfficeChai on Similarweb, 12 October 2026
• AI chatbot market share — Statcounter, September 2026 data
• ChatGPT sends 95% of AI referral traffic — The Next Web on BrightEdge, 24 September 2026
• 3 New Studies Suggest AI Visibility & Search Not Worth Investing In — Search Engine Roundtable, 22 October 2026
• GEO is a distraction — Marie Haynes, 21 October 2026
• ChatGPT vs Google search volume — seoClarity, 28 October 2026
• Search Share Fell To 12% For North American Publishers: Chartbeat — Search Engine Journal, 20 October 2026
• OpenAI Drops UTM Parameters From ChatGPT Results (GPT-6) — Search Engine Roundtable, 14 October 2026, updated 15 October
• ChatGPT's missing UTM tags: what our research shows — Profound, 14 October 2026
• ChatGPT Traffic Drop Appears To Be A Tracking Problem — Search Engine Journal, 15 October 2026
• Google Gemini Adds UTM Parameters For Referral Attribution — Search Engine Journal, 1 October 2026
• Gemini UTM parameters gone — Search Engine Roundtable, 30 October 2026
• OpenAI Tests New Product Labels On ChatGPT Results — Search Engine Roundtable, 6 October 2026
• More ways to connect with the web from Search — Google, 15 October 2026
• Google's Universal Cart reaches US shoppers as Meta flips browsing on — PPC Land, 16 October 2026
• Google AI Overview Citation Cards With AI-Generated Summary Snippets — Search Engine Roundtable, 23 October 2026
• OpenAI To Test Visual ChatGPT Ad Format During AI Image Generation — Search Engine Roundtable, 5 October 2026
• OpenAI ChatGPT Ads Account Manager Representatives — Search Engine Roundtable, 14 October 2026
• ChatGPT lets advertisers target you based on your personal info: report — MobileSyrup, on 404 Media's investigation, 16 October 2026
• OpenAI Auto Enabling Creative Enhancements In ChatGPT Ads — Search Engine Roundtable, 20 October 2026
• OpenAI to rewrite ChatGPT ads by default as advertisers get days to opt out — PPC Land, 21 October 2026
• Inside ChatGPT's Ad Auction: How Relevance, Bids, And Memory Decide Placement — Search Engine Journal, 22 October 2026
• ChatGPT Ads With Learn More Arrow Link; Some Served Through Amazon DSP — Search Engine Roundtable, 26 October 2026
• Google Tests Ads in AI Mode for Personal Intelligence Users — The Keyword, 17 October 2026
• Connect your Google content apps to get a Search services experience that's tailored to you — Google Search Help
• Google privacy policy: connected apps and AI Mode ads — Search Engine Journal, 27 October 2026
• Ads in AI Mode are expanding globally — Google, 28 October 2026
• Google: AI Mode ads earn less — Search Engine Roundtable, 29 October 2026
• Google's ad growth slows to 10.5% as AI Mode ads go to 10 more markets — PPC Land, 29 October 2026
• Google AI Mode tests sponsored and all products together — Search Engine Roundtable, 29 October 2026
• Google Is Fighting SERP Tracking. AI Agents May Make It Worse — Search Engine Journal, 5 October 2026
• Investigating unintended model actions — Anthropic, 9 October 2026
• Google: AI made bot impersonation worse — Search Engine Roundtable, 30 October 2026
• ChatGPT's Most-Cited Sites Tilt Toward Primary Sources, Data Shows — Search Engine Journal on Ahrefs, 12 October 2026
• Understanding ChatGPT Citations: Evidence from Production Web Search Across Model Generations — arXiv preprint, 11 October 2026
• OpenAI's API Only Partly Matched ChatGPT's Citations, Paper Finds — Search Engine Journal, 18 October 2026
• Cloudflare and OpenAI: making AI discovery better for everyone — Cloudflare, 15 October 2026
• From Public Posts to AI-Search Citations: Measuring the Fragility of AI Search — arXiv preprint, 8 October 2026
• AI Search Tools Cited A Made-Up Concept Within A Week, Paper Finds — Search Engine Journal, 19 October 2026
• Parametric knowledge optimization — Graphite, 16 October 2026
• Some Brands On ChatGPT's Cited Pages Still Got Few Mentions — Search Engine Journal, 26 October 2026
• Citation Worthiness: Which GEO Tactics Hold Up Across 96,418 URLs — SISTRIX, 22 October 2026
• AI answer engines often bypass local news outlets — Press Gazette, 22 October 2026
• 2026 AI citation study — TJ Digital, 25 October 2026
• Search Engine Journal's write-up of the TJ Digital study — Search Engine Journal, 28 October 2026
• What is AI reading? October 2026 — Muck Rack, last updated 28 October 2026
• The half-life of an AI citation is 11 days — Profound, 30 September 2026
• How stable are AI citations? — Ahrefs, 29 October 2026
• Measuring AI visibility — SISTRIX, 28 October 2026
• Topics Portfolio (beta) in AI Performance — Bing Webmaster Blog, 13 October 2026
• Bing Webmaster Tools Sorts AI Search Queries Into Topics — Search Engine Journal, 13 October 2026
• Meta's New Terms Expand Crawling Rights Beyond Advertisers — Search Engine Journal, 16 October 2026
• Meta Drops Training From Its Crawl Clause And Keeps The Crawl — Search Engine Journal, 19 October 2026
• Meta Delays The Terms That Let It Crawl Business Sites — Search Engine Journal, 26 October 2026
• Google says robots.txt limits AI crawling, not what bots do with data — PPC Land, 21 October 2026
• Judge dismisses Chegg's and Penske's AI Overviews cases: an expectation is not an agreement — The Decoder, 4 October 2026
• Penske AI Overviews lawsuit dismissed because 'no formal bargain' struck with Google — Press Gazette, 1 October 2026
• 400 publications are now partnering with us on our News AI pilot program. — Google, 15 October 2026
• Reach CEO touts Google AI deal as a template, steps up pressure on OpenAI over scraping — Digiday, 16 October 2026
• News publisher AI deals and lawsuits tracker — Press Gazette, updated 12 October 2026
• SPUR publishes AI content tracking standard — Digiday, 2 October 2026
• Usage Policy — Anthropic, effective 12 November 2026
• Claude's New Rules Target Fake Sources Built To Sway AI Answers — Search Engine Journal, 9 October 2026
• Google Adds Fake Author Warning To Helpful Content Guidance — Search Engine Journal, 2 October 2026
• CMA strengthens proposals on search choice — UK CMA, 23 September 2026
• EU AI Act update: the Digital Omnibus — Orrick, 29 July 2026