AI trust deficit is now a revenue line, Usercentrics data shows
Consent infrastructure has graduated from compliance overhead to competitive moat. That is the headline finding from Usercentrics' State of Digital Trust 2026, a Sapio Research survey of 11,000 consumers across the UK, US, Germany, Netherlands, Sweden, Spain, and Italy, which maps how concern about AI data handling has hardened into measurable purchasing behaviour.
The numbers are stark. Almost half of respondents, 47%, took at least one action with direct revenue consequence in the past six months because of concerns about how brands were using their data in AI systems. One in four cancelled a subscription outright. One in five switched to a rival they judged more trustworthy. On the other side of the ledger, 52% say they would pay a 7% average premium to a brand that is transparent about its AI data practices. In Germany, willingness to pay that premium reaches 73%, at an average uplift of 9%.
From sentiment to commercial signal
The year-on-year movement underpinning these figures is the sharpest in the dataset. The share of consumers who trust AI less than humans with personal data has risen from 48% to 52%, the largest single shift recorded across the survey's tracking history. Cookie acceptance is declining in parallel: 48% of consumers click "accept all" less often than three years ago, a structural behavioural change that degrades the first-party data signals that digital advertising and personalisation stacks depend on.
Usercentrics strategy lead Tilman Harmeling frames the commercial stakes plainly: "Consumers are making purchasing decisions 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."
One figure that has not moved is equally significant. Despite two years of high-profile AI controversy, 46% of consumers still lack a clear understanding of how their data is collected and used, identical to 2025. The knowledge gap is structural, not cyclical, and the report argues it represents the clearest commercial opportunity available: consumers who understand what is happening with their data are nearly three times more likely to be comfortable with personalisation than those who do not.
The convergence angle: where AI governance meets capital strategy
For cross-sector leaders, the report surfaces a dynamic that reaches well beyond any single brand's consent banner. Agentic AI, the class of systems that now books meetings, accesses inboxes, and connects to financial accounts on users' behalf, has fundamentally changed the consent question. It is no longer about what a user chooses to share; it is about what an AI agent accesses before making a decision. Only 8% of consumers are fully comfortable with AI assistants accessing their data without conditions, and 37% are comfortable with AI access to financial accounts at all, the lowest comfort level of any category tested.
This sits at the intersection of enterprise software strategy, financial services liability, and the emerging regulatory stack. The EU AI Act moved into active enforcement during the study period. More than 20 US states now have comprehensive privacy laws in effect with no federal standard unifying them. In the US specifically, only 39% of consumers trust government services with their data, the lowest of any market surveyed, creating a vacuum that brands can either fill or cede. The finding has direct implications for wealth managers, insurers, and consumer banks deploying agentic tools: the liability profile of "resigned consent" (allowing access because opting out feels harder than complying) is a brand risk sitting off most enterprise risk registers.
For capital allocators, the data reinforces a thesis forming across the privacy-tech and enterprise-software sectors: consent infrastructure is becoming critical business infrastructure, and the firms treating it as such are building defensible pricing power. The 38-point spread in willingness to pay for AI transparency between Germany (73%) and the Netherlands (35%) also signals that European market entry and expansion strategies cannot treat the region as a single trust environment. Geography, generational cohort, and regulatory familiarity are all material variables for any platform pricing AI transparency as a differentiator.
Usercentrics, which processes more than 8.8 billion user consents monthly across 2.4 million websites, publishes the full report alongside its T.R.U.S.T. framework for building consent infrastructure ahead of the next phase of agentic AI deployment.