Three Verticals, One Signal: The Generative-AI Wedge Entering the Corpus, 2024-2026

One signal, three domains

Three separate research programs, run by three separate organizations for three unrelated purposes, each turned up the same finding in roughly the same period: consumers are starting to use generative AI to decide where to go and who to hire.

• News: 7% of respondents used AI chatbots weekly for news in 2025, rising to 15% among adults under twenty-five. (Reuters Institute, Digital News Report 2025)
• Dining: 44% of Americans say they plan to use AI more for restaurant discovery and booking in 2026. (OpenTable, 2026 Dining Trends Report)
• Legal: 14% of consumers have used AI to answer legal questions, rising to 26% among Millennials and 23% among Gen Z; 28% of those users were referred to contact a lawyer. (Clio, Legal Trends Report 2024/2025)

Why convergence matters more than any single number

Read in isolation, each of these figures is thin evidence. A single-year, vendor-commissioned survey with partial methodology disclosure is a weak basis for a strategic claim on its own; that is a fair objection to any one of the three numbers above.

What changes the picture is that news, dining, and legal share no common survey house, no common sample, and no common incentive to report the same trend. Reuters Institute studies media consumption. OpenTable studies restaurant bookings. Clio studies legal-practice usage. None of the three organizations coordinated with the others, and none has a reason to inflate a shared narrative. Three independent measurements landing on the same early-adoption pattern, in the same period, is a stronger form of evidence than any one of them individually, even though each remains tiered emerging on its own.

News: the fastest-moving of the three

The Reuters Institute's Digital News Report is a 48-market survey run on YouGov panels with disclosed methodology, an established measurement instrument. The 2025 wave found weekly AI-chatbot use for news at 7% overall and 15% among under-25s, still a minority behavior but doubling the age-cohort rate in a single reading.

That figure sits alongside a related finding from the same research program: social media has become the dominant main news source for 18 to 24 year olds, cited by 44% of that cohort, a reversal from a decade earlier when the same age group named news websites and apps as their primary source. Read together, the two figures describe a younger cohort whose main news channel has already shifted once, toward social and video, and is now showing the first sign of shifting again, toward conversational AI.

Dining: an industry betting its trends report on it

OpenTable's 2026 Dining Trends Report is commissioned consumer research, not an independent academic instrument, and its methodology is only partially disclosed in the coverage available. The 44% figure, Americans planning to use AI more for restaurant discovery and booking in 2026, should be read as a stated intention rather than a measured behavior; intention and behavior diverge in survey research more often than not.

Even discounted for that limitation, the figure is notable for what it signals about industry attention: a hospitality-sector research report chose to lead with an AI-discovery statistic at all, which is itself evidence that the category is starting to take the surface seriously, independent of whether the specific number holds up.

Legal: discovery may not be the real bottleneck

Clio's Legal Trends Report combines the company's own legal-practice-management product usage data with a commissioned consumer survey, a methodology that is partially disclosed. Its headline figures here, 14% of consumers using AI for legal questions, rising to 26% among Millennials and 23% among Gen Z, with 28% of those users referred onward to contact a lawyer, describe the same early-adoption pattern seen in news and dining.

The same research program surfaces a separate, sobering finding worth naming alongside it: in Clio's own contact-intake field study, only 33% of law firms responded to a test email inquiry and 40% answered a phone call, both down from 2019. For legal specifically, the emerging AI-discovery signal may matter less than a longer-standing problem: firms are not answering the inquiries they already generate, discovery is not the bottleneck, response is.

What is not yet known

No unified, methodologically consistent instrument measures AI-assisted discovery across verticals using the same sample, question wording, and year. Reuters Institute, OpenTable, and Clio each measure a different population with a different survey design, and stacking their headline percentages side by side, as this piece does, is a way of reading convergence, not a formal comparison of equivalent numbers.

None of the three sources reports what share of AI-assisted discovery actually converts into a completed booking, purchase, or engagement. The gap is real: this evidence establishes that a new discovery behavior is appearing across unrelated domains in the same window, it does not yet establish how large, durable, or commercially consequential that behavior will become.

How a business tracks this without overreacting

A single-digit-to-teens adoption rate does not justify redirecting a marketing budget wholesale toward AI-answer optimization today. It also does not justify ignoring the surface, because the convergence across three unrelated domains, in the same eighteen-month window, is a stronger early signal than any one industry's number would be alone.

The practical response available to any business is direct measurement: sample the real questions your buyers would ask an AI engine, and record whether and how often you are named in the answer. No published dataset will tell you whether your specific buyers have already made this shift; only reading your own AI-answer visibility will, and reading it now establishes a baseline before the surface, if the pattern holds, becomes more contested and more expensive to enter.