Usercentrics this month published the State of Digital Trust 2026, a study of 11,000 consumers across seven markets conducted by Sapio Research in March 2026. The findings document a structural shift: consumers have moved from expressing concern about AI data handling to taking concrete purchasing actions because of it, with consequences that reach directly into ad campaign measurement and bidding algorithm performance.

The report arrives on June, 2026. Fieldwork was completed in March 2026, accurate to plus or minus 0.9 percent at 95 percent confidence. Seven markets were included: Germany, the United States, the United Kingdom, Spain, Italy, the Netherlands, and Sweden. Sweden joins the study for the first time in 2026; year-on-year comparisons therefore exclude that market.

From sentiment to purchasing behavior

Last year's State of Digital Trust captured consumers at a moment of rising concern. This year, according to Usercentrics, the tipping point has passed. The headline number is stark: according to the report, 47 percent of consumers globally took at least one action with a direct revenue consequence in the six months before fieldwork was conducted, because of concerns about how a brand handled their data in the context of AI. These are not expressions of intent. They are completed transactions, or the absence of them.

The breakdown is specific. According to Usercentrics, 24 percent of consumers - almost one in four - canceled a subscription or stopped purchasing from a brand altogether. Another 20 percent switched to a competitor they believed handled AI data more responsibly. A further 20 percent reduced their spending with the brand in question. And 31 percent warned friends and family or complained publicly. These behaviors do not carry equal commercial weight, but they compound. According to the report, 35 percent of consumers globally took two or more of these actions against the same brand.

The scale implication is significant. A brand with one million customers could, according to Usercentrics' figures, have experienced as many as 240,000 purchase-affecting decisions driven entirely by AI data concerns in a six-month window. And those brands are not necessarily doing anything illegal. Many simply have not communicated how they use customer data in AI systems.

The flip side of the churn data is a pricing power signal. According to Usercentrics, 52 percent of consumers globally will pay more for a brand that is transparent about how it uses AI with their data, at an average premium of 7 percent. Of those, 28 percent would stretch to 10 percent above the regular price, and another 16 percent would pay between 11 and 20 percent more.

Market variation on this metric spans 38 percentage points - the widest spread on any commercial measure in the study. Germany leads at 73 percent, at an average premium of 9 percent. The Netherlands sits lowest at 35 percent. The United States and United Kingdom both reach 50 percent willingness, in line with the global average.

Age is a significant variable. Among 18 to 29 year-olds, 67 percent say they would pay more for AI transparency. Among consumers aged 60 and above, that figure falls to 23 percent. As Tilman Harmeling, Strategy and Market Intelligence at Usercentrics, observed in the report: "Consumers are making purchasing decisions also based on how brands handle their data, and over half are willing to pay more to the ones that get it right. The brands that move first won't just earn the premium. They'll earn a category position that's almost impossible to compete against once it's established."

There is also a gender gap that has received little industry attention. According to Usercentrics, 58 percent of men are willing to pay the AI transparency premium, against 47 percent of women. That 11-point gap suggests consumer attitudes toward AI data risk are not uniform, and that communications designed for one broad audience are likely to underperform with a meaningful segment.

AI personalization: the 34-point gap explained

AI personalization is the flashpoint in the data. According to Usercentrics, 71 percent of consumers find AI-driven personalization intrusive. The Netherlands registers the highest concern at 77 percent. The US and UK are both at 75 percent.

Read in isolation, that figure looks like a structural objection to personalization itself. But the report surfaces a distinction that reframes it. Among privacy-aware consumers - those who understand what data is being collected and why - 53 percent are comfortable with AI personalization. Among privacy-unaware consumers, that figure is 19 percent. A 34-point gap on a single variable: communication.

The same awareness gap that makes consumers more cautious about data-sharing also makes them more receptive to personalization when it is properly explained. Brands that explain what they are doing to the right audience unlock nearly three times the consent, according to the report. The barrier is not technology. It is the absence of a clear explanation.

This connection has direct implications for marketing performance. Personalization budgets running on an uninformed audience may be generating lower conversion rates than the same technology applied to an audience that has been told, in plain language, what it is doing with their data and why.

The report's third major finding is the one with the most immediate operational significance for advertising professionals. Cookie consent rates are declining, and that decline is degrading the quality of the signals that power bidding algorithms on platforms like Google and Meta.

According to Usercentrics, 48 percent of consumers are clicking "accept all" less often than they did three years ago. That figure was 46 percent in last year's report - it is moving in one direction only. Among markets, the Netherlands leads the decline at 57 percent clicking accept all less often. Sweden and the United States are the slowest movers at 42 percent each.

What is replacing passive acceptance is active consideration. Over half of consumers now selectively manage their cookie preferences. Thirty-nine percent accept only necessary cookies; 16 percent customize their settings. The user who clicks "accept all" without reading is becoming the minority.

Two forces are driving this shift simultaneously, and they operate differently. The first is consent fatigue: users worn down by years of cookie banners, clicking through to reach content. The second is what the report describes as a privacy awakening: consumers reading the "more information" link more often, understanding what is being done with their data, and making more deliberate choices. The report notes that these two populations require different responses. Fatigued users can potentially be won back through better banner design. Awakened users require genuine transparency.

The advertising performance consequence is direct. Google and Meta's bidding algorithms run on the consent signals brands feed them. When consent drops, signal quality drops. When signal quality drops, cost per acquisition rises. The report notes a 31-point gap on "accept all" rates between privacy-aware and privacy-unaware consumers: 26 percent of privacy-aware consumers accept all by default, against 57 percent of privacy-unaware consumers. As audiences become more informed - a trajectory the data shows is continuing - the consent rates brands have built their measurement infrastructure around are degrading at the source.

This is not a future risk. PPC Land has documented how publisher and advertiser measurement infrastructure depends on clean, consented data at the signal level, with broken consent flows showing up as unexplained softness in campaign performance before they register as compliance exposure.

The understanding gap that will not close

One finding in the 2026 report stands apart for its persistence. According to Usercentrics, 46 percent of consumers still do not have a good understanding of how their data is collected and used. That figure is identical to 2025. Two years of consent banners, cookie notice updates, and compliance communications have not moved it.

The systems built to explain data use to consumers - banners, policies, notices - are not doing the job they were designed to do. The report cites NordVPN research finding that a typical internet user would need a full working week each month to read the privacy policies of every website they visit. The failure is, as the report explicitly states, a design failure, not a user failure. And it compounds every year it goes unfixed.

The market-level picture adds a further complication. Sweden and Germany, two of the most privacy-conscious markets in the study by reputation and regulatory environment, have the largest comprehension gaps. According to Usercentrics, 56 percent of Swedish consumers and 53 percent of German consumers do not have a good understanding of how their data is collected and used. The US and UK perform best on this measure, both at 41 percent - still nearly half the audience.

What this means in practice is that the audience most likely to pay a premium for AI transparency is often the audience least clear on what is actually happening with their data. In Germany, where 73 percent are willing to pay more for AI transparency at an average 9 percent premium, over half of consumers cannot clearly describe the data processes that produce the personalized experiences they are either rewarding or rejecting. The MIT Technology Review and Usercentrics mapped this dynamic in April 2026, arguing that privacy-led UX has become a prerequisite for AI growth rather than a constraint on it.

Brands that explain data practices in plain language - not legal copy - have a structural advantage in markets where high concern coexists with low understanding. Germany, on the data in this report, is the clearest example.

The final finding in the report is described by Usercentrics as potentially the most dangerous because it is the one that looks fine from inside a dashboard.

According to the report, only 8 percent of consumers are fully comfortable with AI accessing their personal data without conditions. A further 23 percent will allow it, but only if they can approve each individual request. Seventeen percent - the resigned consent segment - are uncomfortable with AI accessing their personal data, but would allow it anyway.

Resigned consent appears as compliance in a CRM and as a retained customer in revenue reporting. It is neither. It represents a user who gave up trying to say no: who could not find the decline option, found it too confusing, or decided it was not worth the effort. That user will leave the moment an alternative brand makes the choice simpler. And while they remain, they degrade the quality of signals flowing to ad platform algorithms. Resigned opt-ins train smart bidding systems on signals that do not reflect genuine intent. Lookalike audiences built from them carry that distortion forward.

Comfort with AI access follows a hierarchy of stakes. According to Usercentrics, 49 percent of consumers are comfortable with AI assistants accessing work tools - the highest of any category in the study. Only 37 percent are comfortable with access to financial accounts, the lowest figure. Two-thirds (65 percent) prefer human interaction for healthcare questions; 61 percent for financial advice; 61 percent for handling complaints. These preferences are not, the report argues, evidence of permanent limits on AI. They are a map of where trust infrastructure needs to be built before AI can expand into those contexts.

This connects to a broader industry shift that PPC Land has tracked through 2025 and 2026: agentic AI has moved from a capability that consumers interact with to one that acts on their behalf, booking meetings, accessing inboxes, connecting to financial services, and making decisions within the tools consumers already use. That shift changed the nature of the consent question entirely. Usercentrics acquired MCP Manager in January 2026 specifically to extend consent management into AI-driven workflows, becoming the first major privacy platform to extend consent governance into that layer.

On March 31, 2026, four UK regulators jointly published a foresight paper mapping agentic AI governance requirements for the advertising and marketing sector. The EU AI Act moved from phased implementation into active enforcement. More than twenty US states now have comprehensive privacy laws in effect with no federal standard unifying them.

Market profiles: seven different commercial stories

The report's fifth chapter breaks down findings by market. Three broad patterns emerge across the seven countries.

Where concern has converted into action: Spain and Germany lead, with action rates of 76 percent and 75 percent respectively against brands over AI data concerns, both well above the 67 percent global average. In Germany, concern long ago became behavior - and the commercial reward for transparency is the highest in the study.

Where institutional trust has collapsed: The United States, where only 39 percent of consumers trust government services with their data - the lowest of any market in the study. Half of American consumers will pay more for a brand that is transparent about AI, and 78 percent would stop using a service over data misuse. The absence of a regulatory floor comparable to the EU's General Data Protection Regulation means every point of trust a US brand earns, it earns on its own terms.

Where high concern has not yet translated into action: The Netherlands ranks 77 percent on AI personalization concern - the highest in the study - yet only 53 percent have taken action against a brand and just 35 percent are willing to pay more for AI transparency. That pattern of high anxiety and low activation suggests either a population that has not found a framework for translating concern into action, or brands that have not yet given them a compelling reason to.

The United Kingdom is moving fastest along the awareness curve. According to the report, the share of UK consumers unaware of their data privacy rights fell by seven points in a single year, from 50 percent in 2025 to 43 percent in 2026 - the largest single-market rights awareness movement in the study. And 80 percent of UK consumers say they would stop using a service if their data was misused, the highest threshold of any market.

Italy sits close to global averages on most dimensions. The outlier is willingness-to-pay intensity: 42 percent of Italian consumers are willing to pay more for AI transparency, but the average premium is just 5 percent, the lowest of any market. Trust functions defensively there rather than as an offensive pricing tool.

Sweden enters the study for the first time. The baseline numbers contain an unusual combination: 69 percent trust banking with their data, the highest banking trust figure in the study; 56 percent do not understand how their data is collected and used, the highest comprehension gap of any market, ten points above the global average; and 60 percent find AI personalization intrusive, the lowest concern figure in the study.

What the T.R.U.S.T. framework prescribes

The report's sixth chapter organizes the practical response into a five-step sequence Usercentrics calls the T.R.U.S.T. Framework. The steps are Translate, Remove, Unify, Secure, and Track. The sequence is deliberate. Each step builds on the one before it, and the report notes that most organizations fail by jumping ahead before earlier steps are solid.

Translate means getting the consent moment itself right: a banner written for the person reading it, not for the legal team that approved it. Remove means auditing what is getting in the way of an honest user choice - giving equal weight to accept, decline, and customize, and ensuring controls are reachable in two clicks. Unify means extending the standard across every consent-relevant touchpoint so that the banner, the preference center, the data subject access request tool, and AI disclosures all feel like the same brand making the same promise. Secure means mapping where consent signals go and ensuring AI tools do not become shadow data processors outside the visibility of governance teams. Track means measuring trust as a commercial signal rather than an opt-in rate - using retention, churn, complaints, and data subject access request volume as the operational metrics.

The report offers a self-assessment diagnostic - five questions that locate an organization in one of three tiers. Organizations in Tier 1 have a banner and a policy and are broadly compliant, but do not know their current consent rate and have not benchmarked it. Tier 2 organizations have measured consent rate and fixed obvious dark patterns, but the broader experience feels disconnected - different touchpoints feel like different brands. Tier 3 organizations have strong consent infrastructure and governed data flows but no strategy yet for the specific risk of AI agents acting on customers' behalf.

PPC Land has tracked how consent infrastructure has become the central variable in first-party data quality across multiple 2025 and 2026 studies, with the connection between consent rate and campaign signal quality becoming more direct as bidding algorithms have grown more dependent on consented behavioral data.

Why this matters for the marketing community

The State of Digital Trust 2026 is relevant to digital marketers for a reason that sits upstream of brand reputation management: consent rate is a performance measurement variable.

When a consumer declines consent, the behavioral signal that would have fed smart bidding, lookalike audience construction, and attribution modeling does not exist. When they provide resigned consent - allowing tracking because opting out is too difficult - the signal that reaches the algorithm may not reflect genuine commercial intent. The quality difference between these two states is not visible at banner level or in a consent rate dashboard. It surfaces as unexplained cost-per-acquisition drift, weakened lookalike audience performance, and gradually degrading attribution coverage.

Previous PPC Land coverage of the 2025 State of Digital Trust found 46 percent of consumers at that point accepting cookies less frequently than three years before. The 2026 figure is 48 percent. The direction has not changed. The IAB has positioned consent management as the foundational infrastructure layer for first-party data strategies, not a downstream compliance function - and the Usercentrics data gives that argument quantitative backing at the consumer behavior level.

Multiple research releases tracked by PPC Land through early 2026 have documented that AI adoption is accelerating while the trust infrastructure supporting it remains thin. The Usercentrics report adds the first large-scale, multi-market measure of what that gap costs commercially: 240,000 purchase-affecting decisions for every million customers, and a 7 percent pricing premium available to brands that close it.

Timeline

  • May 2025 - Sapio Research conducts fieldwork for the State of Digital Trust 2025 report on behalf of Usercentrics, surveying 10,000 consumers. PPC Land coverage
  • July 1, 2025 - Usercentrics publishes the State of Digital Trust 2025 report. PPC Land coverage
  • October 15, 2025 - Usercentrics surpasses EUR 100 million in annual recurring revenue, processing more than 7 billion consent decisions monthly across 2.3 million websites and apps. PPC Land coverage
  • December 2025 - A Verve survey finds 65 percent of consumers worried about AI data training, with 97 percent demanding greater transparency from publishers. PPC Land coverage
  • January 14, 2026 - Usercentrics acquires MCP Manager, extending consent management into AI-driven workflows via the Model Context Protocol. PPC Land coverage
  • February 2026 - Spain's data protection authority AEPD publishes a 71-page guide on agentic AI and GDPR compliance. PPC Land coverage
  • February 25, 2026 - France's CNIL opens public consultation on draft recommendation governing session replay tools, requiring prior user consent. PPC Land coverage
  • March 2026 - Sapio Research conducts fieldwork for the State of Digital Trust 2026 across 11,000 consumers in seven markets.
  • March 3, 2026 - Shift Browser survey of 1,448 Americans finds 81 percent concerned about AI data access and 32 percent using AI daily. PPC Land coverage
  • March 31, 2026 - Four UK regulators jointly publish a foresight paper mapping agentic AI governance requirements for the advertising and marketing sector.
  • April 2, 2026 - Cloaked survey finds only 18 percent of Americans trust AI to keep their personal data secure. PPC Land coverage
  • April 14, 2026 - Independent audit of 7,000+ California websites finds Google, Meta, and Microsoft setting ad cookies after users opt out. PPC Land coverage
  • April 17, 2026 - MIT Technology Review and Usercentrics publish report arguing privacy-led UX is a prerequisite for AI growth. PPC Land coverage
  • May 2026 - Iubenda publishes guide on shifting European cookie consent rules and their implications for marketing performance. PPC Land coverage
  • June 24, 2026 - Usercentrics publishes launch video for the State of Digital Trust 2026 on YouTube.
  • June 27, 2026 - Usercentrics publishes the State of Digital Trust 2026 report, with fieldwork from March 2026 across 11,000 consumers in seven markets.

Summary

Who: Usercentrics, a Munich-based data privacy technology company that processes more than 8.8 billion user consents monthly across 2.4 million websites and apps in 195 countries, commissioned the research. Fieldwork was conducted by Sapio Research.

What: The State of Digital Trust 2026 is the second annual study examining consumer attitudes toward data privacy, AI handling, and brand accountability. It surveyed 11,000 consumers across Germany, the United States, the United Kingdom, Spain, Italy, the Netherlands, and Sweden on topics including consent behavior, AI personalization comfort, willingness to pay more for transparent brands, and agentic AI access preferences.

When: Fieldwork was conducted in March 2026. The report and accompanying launch video were published on June 27, 2026. Year-on-year comparisons reference the State of Digital Trust 2025, which was published on July 1, 2025.

Where: The study covers seven markets: Germany, the United States, the United Kingdom, Spain, Italy, the Netherlands, and Sweden. Sweden joins the study for the first time in 2026. The findings are global in scope but show wide variation by market, with Germany showing the highest willingness to pay for AI transparency at 73 percent and the Netherlands showing the highest concern about AI personalization at 77 percent.

Why: Consumer concern about AI data handling has crossed a threshold into purchasing behavior, with 47 percent of respondents globally taking at least one action with a direct revenue consequence in the six months before fieldwork was conducted. The report aims to quantify the commercial cost of the trust gap and provide marketing teams with a structured framework for closing it, given that low-quality and resigned consent signals are degrading the performance of the ad platform algorithms that most digital campaigns depend on.