AI ad labels cut clicks 31.5% as IAB issues disclosure framework

The Interactive Advertising Bureau published the second version of its AI Transparency and Disclosure Framework on August 18, 2026, and buried inside it sits a number that will shape how the rest of the year goes for creative teams. Research from NYU Stern School of Business, cited in the framework and reported by PPC Land on August 19, found that adding an AI disclosure to an advertisement cut click-through rate by 31.5%.

That figure lands in an industry that has spent eighteen months moving in the opposite direction. Some 91% of advertisers report using or planning to use AI for creative production. The share of advertising executives reporting AI deployment reached 83% in 2026, up from 60% in 2024. Cadbury generated 130,000 unique AI videos for a single campaign. The machinery is built. The label is the tax.

What makes the framework awkward is the gap it documents between what advertisers believe and what audiences report. Some 82% of advertisers think consumers feel positive about AI in advertising. Among Gen Z and Millennial respondents, 45% do. That is a 37-point spread, and it runs in the direction that costs money. Separately, 76% of US adults said distinguishing AI content from human content is extremely or very important to them, while 73% of Gen Z and Millennial respondents said knowing AI was involved would not reduce their likelihood of purchase. Those two findings sit uneasily together, and the framework does not resolve them.

Caroline Gigerich, the IAB's vice president for AI and the working group lead, framed the design problem in plain terms. "Not every use of AI needs a label - labeling everything teaches consumers to ignore labels," she said. "Under-disclosure leaves consumers at risk of being misled. Over-disclosure could negatively impact advertisers." Her broader argument rested on a simpler claim: "Trust is everything between a brand and its customers, and being honest about AI is part of earning it." David Cohen, the IAB's chief executive, called trust "a foundational element critical to the growth of AI across the ecosystem."

Where the line falls

Version two spends most of its length drawing that line. Disclosure triggers include AI-generated images and video produced from prompts, synthetic voices of deceased persons regardless of whether the estate authorized the use, synthetic voices of living persons placed in fabricated scenarios, photorealistic AI influencers, digital twins of the deceased, digital twins of living people in events that did not happen, and conversational agents, which get an "AI-powered" label rather than an "AI-generated" one.

The exemption list is longer and more consequential. Routine post-production stays unlabeled: colour correction, dust removal, upscaling. So do stylized and fantastical imagery, authorized voice clones used for commercial endorsement, generic synthetic voiceovers with no identifiable person behind them, background music, audio processing, text and copy generation, and animated characters or mascots. Copy generation being exempt matters more than it sounds, given that text is where most production AI spend currently sits.

Implementation is specified down to the glyph. The framework nominates the Unicode sparkle character, U+2728, or the plain text string "AI-generated," rendered at a contrast ratio of at least 4.5:1 to meet WCAG AA. Video labels belong in the first frame. Audio labels go before the segment begins. Underneath the visible layer sits a metadata layer built on Coalition for Content Provenance and Authenticity credentials, with two custom assertions, com.iab.threshold and com.iab.disclosure, carrying tool identification and timestamps through the supply chain.

The rollout schedule runs 24 months. Companies get 60 days to designate an AI Disclosure Lead. Months zero to six cover team training and a pre-launch checklist. Months six to twelve are for controlled pilots that measure the impact of disclosure on performance, which is where the 31.5% figure will either be confirmed or contradicted at scale. Months twelve to twenty-four are for scaling and automation.

Voluntary rules inside a legal deadline that already passed

The framework is not law, and it arrived after the law did. Article 50 of the EU AI Act took effect on August 2, 2026, alongside the California AI Transparency Act, and the European Commission adopted its Article 50 Guidelines on July 20. Roughly 190 companies had signed the EU Code of Practice by the end of July. Penalties under Article 50 reach 15 million euros or 3% of global turnover, a structure PPC Land covered when EU AI content rules landed on publishers and again when Brussels released free labelling icons ahead of the deadline.

New York's synthetic performer law entered force on June 9, 2026, with fines of $1,000 for a first violation and $5,000 for subsequent ones, a combination PPC Land examined alongside the European penalties. California's statute is optional in form and applies to providers with more than one million users. South Korea granted a one-year grace period and has stricter amendments pending. China's labeling requirement has been live since September 1, 2025. Vietnam's implementing decree took effect in May 2026. India's IT Rules mandate synthetic content labeling.

Graham Wilkinson, executive vice president and chief innovation officer at Acxiom, put the compliance map bluntly: "Regulators in New York, California, South Korea and the EU are setting enforceable AI disclosure rules." Simon Poulton, executive vice president for innovation and growth at Tinuiti, described the practical handoff: "Agencies translate this framework into what runs in-market for clients." Michael Lampert, senior director at Mondelez International, offered the buy-side reading: "Consumers are open to trusting AI in advertising, but only when brands are upfront about it."

There is one divergence worth flagging, because it will produce inconsistent creative across markets. The IAB framework exempts authorized voice clones and authorized digital twins used in endorsements. Article 50 requires labeling regardless of authorization and regardless of whether a consumer would expect it. An advertiser following the voluntary standard in full would still be short of the binding European rule. Platform-level liability has already been pushed downstream: Google assigned AI labeling responsibility to advertisers rather than absorbing it, and DV360's API gained an AI attestation field ten days before the August 2 deadline. Meta and TikTok are named in the framework alongside Google, with Digitas, Conde Nast and Warner Brothers Discovery among the working group participants.

Nobody can prove the agent saw the ad

The measurement question that the label debate assumes has an answer, does not. A Digiday Media Briefing published August 20 laid out the structural problem with advertising placed for machine consumption: the agent retrieves the ad, folds it into a generated answer, and may shape a purchase decision without the person ever seeing the original creative. There is no impression in the conventional sense, and no click.

Brendan Norman, chief executive of Classify, reduced it to a single question: "How do you attribute that the ad was actually seen by an agent and then influenced the response?"

Three approaches are being tested, and none has become a standard. OpenAds embeds unique ten-digit referral codes inside ad copy. Steven Liss, the company's co-founder, described tracking whether the code survives the retrieval process when ChatGPT scrapes the page, then whether a user eventually clicks through. His commercial framing was direct: "For advertisers to put anything more than experimental budgets in, they're going to have to be able to measure some kind of outcome."

Oasy takes a delivery-side route, inserting dynamic text-based ads into HTML at the moment an AI crawler requests a page, using CDN integrations including Cloudflare. Co-founder Choy Travers was candid about where that leaves attribution: "Within AI visibility there is no exact science. There's no exact attribution, unlike Google Ads and Facebook Ads, where you have that click." Oasy is working toward cost-per-click or cost-per-referral pricing through partnerships with native ad platforms.

Time magazine has taken a third path, publishing FAQ-style sponsored units inside markdown pages and then measuring visibility through tools such as Mobian, which submits questions to AI search engines and returns brand visibility, favourability and accuracy scores. Time's revenue mix explains some of the urgency: events are expected to supply more than half of its revenue in 2026, against 28% in 2023.

Two structural numbers frame the whole exercise. Roughly 26% of ChatGPT responses now carry sponsored ads. And average robots.txt coverage of AI bot rules across publishers sits at about 21%, meaning the majority of the crawlable web has no explicit position on the agents reading it.

The one-day window

Against that backdrop, OpenAI's own measurement layer took a small step forward. ChatGPT Ads gained view-through conversions with a fixed one-day window, documented on August 18 and 19 and reported by PPC Land on August 19. The metric appears as VTA (1d) at campaign, ad group and ad level in Ads Manager. The window is not configurable. It operates independently of click-through settings. When a click and an impression both qualify for the same conversion, the click takes precedence.

What the column does not do is more interesting than what it does. View-through conversions are excluded from the headline Conversions total, from cost-per-acquisition calculations, from post-click conversion rate, from bidding, from billing, and from conversion optimisation. OpenAI describes the data as supplemental reporting that "does not feed any of those systems." That is a conservative choice, and it stands in contrast to the rest of the industry, where view-through credit routinely flows into optimisation and invoices.

The build-out has been quick. The advertising pilot launched February 9, 2026, with no conversion measurement at all. The Measurement Pixel and Conversions API arrived May 5 alongside a self-serve Ads Manager with CPC bidding. Conversion-optimised campaigns began rolling out June 5. Automatic advanced matching switched on by default on August 17. Six months, from no measurement to a partial view-through column that is deliberately walled off from the auction.

Television reaches for older instruments

While the AI surfaces argue about whether an impression exists, the incumbent currency is being retooled. Nielsen said its Big Data + Panel video measurement would ship upgrades by the end of August, AdExchanger reported on August 19, timed ahead of the September television season.

Three changes matter. Demographic and Spanish-language audience estimates will draw on the American Community Survey and the National Hispanic Enumeration Survey, and will key on the primary language spoken in a household rather than on surname. Brian Fuhrer, Nielsen's senior vice president for product strategy and thought leadership, said primary household language has a more significant impact on viewing patterns than the surname proxy it replaces.

Co-viewing measurement will lean harder on wearable devices Nielsen introduced in 2021, a passive approach that avoids asking panel members to log in. The method was piloted during the Super Bowl earlier in 2026. And the ARF DASH TV universe study data, integrated at the start of the year and met with publisher objections, is being refreshed more frequently. Fuhrer described the previous position as a long lag time in a rapidly changing environment, with figures running twelve to fifteen months old. That data now feeds the Gauge report, which ranks content by watch hours.

The through-line across all three of these stories is that measurement is being rebuilt in two places at once, on entirely different timescales: a panel-plus-ACR currency correcting a fifteen-month lag, and a set of startups arguing over whether a ten-digit code inside an ad can survive a language model's retrieval step.

Two agents, one MCP server, and a $10,000 test

The clearest evidence this week that agentic buying has left the demo stage came from a small independent agency and a radio company. Butler/Till, based in Rochester, New York, extended its agentic media buying tests into audio with iHeartMedia, Digiday reported on August 20.

The mechanics are worth spelling out. Two AI agents were paired, one built by Butler/Till and one by iHeartMedia, communicating over an MCP server. A human planner supplied the brief. The buying agent then negotiated directly with the publisher's agent. The campaign ran roughly four weeks across July and August 2026 for an undisclosed US agricultural client, against just under $10,000 in media, buying streaming audio and podcast inventory.

Two results came back. CPMs came in 42% below the client's direct-buying benchmark. And 48% of podcast impressions landed in premium non-skippable mid-roll positions, against 33% under the traditional approach. The second number complicates the first: cheaper inventory that skews toward better placements is not the usual trade, and a four-week test at that budget cannot settle whether the effect holds at scale.

The architectural shift is the more durable detail. Butler/Till's earlier pilots, run across a dozen clients on CTV and digital display, routed through SSP intermediaries including PubMatic. This one did not. Agency agent talked to publisher agent, with the exchange layer removed from the path. Kristie Murphy, associate director for programmatic at Butler/Till, said the work is "moving away from the one-off test phase" and that "there's more potential for agentic beyond one channel." Scott Ensign, the agency's chief strategy officer, described the governance model: "AI can execute media buying within human-defined strategy and governance, creating a faster marketplace." Lisa Coffey, chief business officer at iHeartMedia, gave the supply-side motive: "Agentic media buying is happening because it's getting audio a much bigger feature in the digital ecosystem."

MCP quietly becomes plumbing

The same protocol turned up in an entirely different context two days earlier. Microsoft Clarity published five MCP query patterns for behavioural analytics on August 18, showing how natural-language requests map onto session data: pages with scroll depth above 75% but conversion below 5%; sessions combining rage clicks, excessive scrolling and quick backs; engagement time segmented by device, channel, region and browser; pages whose session duration moved more than 20% month over month; and session recordings pulled for pages with low scroll depth and high exit rate.

The limits are tight. Ten requests per day per project. A three-day maximum historical window per request. Three dimensions per query. Node.js v16 or later, an active Clarity project, and a Data Export API token, with Claude for Desktop and Cursor named among compatible clients. PPC Land also noted that the sample tables in Microsoft's post are illustrative placeholders rather than measured findings, which is the sort of caveat that tends to get lost in the retelling.

The chronology shows how fast this layer has assembled. Anthropic introduced the Model Context Protocol in November 2024. Clarity's MCP server launched June 4, 2025. AI channel groups separating ChatGPT, Claude, Gemini, Copilot and Perplexity arrived August 29, 2025. Clarity Citations reached general availability May 13, 2026. Web IQ, citation reporting and a Microsoft Advertising MCP pilot were announced at Cannes Lions on June 17, 2026. Topic Insights launched July 9, Query Topics entered beta July 22, and an AI Scrape-to-Referral Ratio card was added August 13. Twenty-one months from protocol announcement to a buying agent negotiating a podcast rate over the same standard.

Microsoft ships its automation bundle, on different terms

Microsoft Advertising rolled out AI Max across all accounts globally on August 19, announced by ads liaison Navah Hopkins on LinkedIn. The bundle contains three features: search term matching, which expands query eligibility beyond keyword lists using keywords, ads, landing pages, intent and contextual signals; text customization, which generates messaging variations from existing assets and site content and picks combinations at auction time; and final URL expansion, which routes visitors to whichever page matches their intent, with exclusion controls.

Hopkins said term matching helps advertisers "appear on and convert more complex conversational queries, especially ones in AI experiences on Bing and Copilot."

Controls shipped at launch rather than being retrofitted. Each campaign supports up to 20 brand lists, a cap that applies to lists rather than to individual brands. Brand exclusions keep ads off queries naming specified brands. Brand inclusions define which branded queries a campaign may target. Term exclusions block named phrases from entering machine-written copy.

The timing invites comparison. Microsoft moved AI Max to open pilot on June 17, 2026, confirmed Predictive Matching's retirement into AI Max on July 28 and 29, and went global on August 19 as an opt-in. Google's equivalent conversion is scheduled for September 1 and is not optional. A separate consideration sits underneath both: a Lunio study published in August 2026 found Google AI Max campaigns exposed to 72% more invalid traffic than standard search, with AI Max accounting for 68% of all invalid clicks detected across roughly 414 million clicks between October 2025 and June 2026.

The price everyone sees, and the price they actually get

The Federal Trade Commission voted 2-0 to authorize a proposed enforcement policy statement declaring that undisclosed personalized pricing is likely to constitute an unfair or deceptive act or practice under Section 5 of the FTC Act, PPC Land reported on August 19. A 30-day public comment period begins on Federal Register publication.

Chairman Andrew Ferguson stated the premise: "When consumers see a listed price, they expect it to be same price that everyone else sees."

The statement runs both the deception test and the unfairness test at the same target, and sets out seven illustrative scenarios of problematic practice. It draws support from the Fair Credit Reporting Act and from state insurance laws in California, Virginia and Washington, which already restrict what data may influence an individual quote.

The company names attached to the underlying 6(b) study are a map of the pricing stack: Mastercard, Revionics, Bloomreach, JPMorgan Chase, Task Software, PROS, Accenture and McKinsey. PPC Land covered that probe when it opened, and separately examined how shopping AI intensified the surveillance pricing debate. Prior enforcement actions against StubHub, Instacart, Greystar and Hopper supply the precedent. Named in the current document as adjacent capabilities: Amazon's Dynamic TV Creative, Microsoft Advertising's work with Epsilon, and Google's Universal Commerce Protocol.

For advertising specifically, the significance is that personalization infrastructure and pricing infrastructure have converged. A creative that varies by audience segment is settled practice. A price that varies by the same segment, without disclosure, is now positioned as a Section 5 problem.

Target cuts prices with a $994 million refund

The commercial counterpoint arrived the same week. Target reported second-quarter results on August 19, and the numbers were reshaped by a single line item: $994 million in pre-tax tariff refunds. Digiday reported on August 20 that the retailer intends to push that money into lower shelf prices.

Net sales rose 5.3% to $26.5 billion. Operating income reached $2.6 billion, a 94% increase, at an operating margin of 9.6% of sales. Net profit came in at $1.87 billion, more than double the prior-year figure, with earnings per share above $4. Full-year operating margin guidance improved by roughly 90 basis points to 6%. Capital expenditure in the quarter was $1.4 billion, up 27%.

The advertising line grew faster than the business around it. Adweek reported on August 19 that Roundel revenue rose 28.6% year over year to $279 million, against $217 million in the same quarter of 2025. Non-merchandise sales, which include Roundel, loyalty and marketplace, rose 20.1%. Comparable store traffic was up 3.6%, digital sales up 8.7%, and same-day delivery grew more than 25%. Wish list creation rose 50%.

Chief financial officer Jim Lee tied the refund to pricing directly: "We have, and we will continue to, invest in price to ensure our guests are getting tremendous value each and every time they visit us at Target." On the tariffs themselves, Lee added: "We continue to look first to find other means to reduce tariff impact, including changes to country of origin and vendor collaboration." Chief executive Michael Fiddelke called the quarter "an important step forward in the plan we laid out earlier this year to open a new chapter of growth for Target." Chief merchandising officer Cara Sylvester reported that items added to lists more than doubled and that conversion across key back-to-school pages rose nearly 20%.

Target has already reduced prices on more than 10,000 items annually and has committed to further cuts. Back-to-school items were priced below the prior year, half the assortment was new, and exclusive items were held under $25. All six core merchandising categories improved year over year, with double-digit growth in toys, electronics and books, and high single-digit growth in food, beverage and beauty.

Other retailers made comparable moves with their refunds. Walmart put the money into lower prices. BJ's Wholesale cut prices by roughly 0.5%. Amazon issued direct refunds in limited circumstances.

Read together with the FTC statement, the sequence is instructive. A retailer that has just published a 28.6% increase in advertising revenue and a doubling of profit is choosing to advertise price reductions in public, in the same week that the regulator moved against prices that are not public at all. The visible price is becoming a marketing asset precisely as the invisible one attracts legal risk.

Also noted

• August 19, 2026 - Taboola's Realize platform took over global programmatic display sales on NBCNews.com and TODAY.com, the company's first deal beyond native inventory, announced by chief executive Adam Singolda and NBCUniversal News Group chief business officer Satpal Brainch. PPC Land
• August 19, 2026 - Google Ads customer retention settings gained dedicated toggles for bidding higher on loyalty program members and surfacing member benefits in eligible countries, spotted by Arpan Banerjee on LinkedIn. Search Engine Roundtable
• August 19, 2026 - Google Trends began rolling out interactive Explore maps and a regional breakdown view for geographic trend analysis, announced on LinkedIn and available at trends.google.com/explore. Search Engine Roundtable
• August 19, 2026 - DiVine, a six-second video app backed by Jack Dorsey and modelled on Vine, launched August 20 with a strict no-AI-content policy and left the door open to advertising as a future revenue line. Digiday
• August 19, 2026 - The Atlantic launched Rabbithole, a daily trivia game whose seven interconnected questions are written by hand each day by Jeopardy champion Drew Goins from that day's articles. Digiday