45% of consumers now ask AI tools for local business recommendations, up from 6% one year earlier (BrightLocal, Local Consumer Review Survey 2026). That shift — alongside ChatGPT's 900 million weekly active users (OpenAI, February 2026, via TechCrunch) and Bain & Company's finding that roughly 60% of searches end without a click — means the first screen many buyers see is now a written answer, not a list of links.
This review compiles what published, named-source research says about how buyers use AI assistants to choose businesses in 2026, what behavior changes inside an AI answer, and which content attributes measurably increase the odds of being recommended. Every figure below carries its source inline. Where evidence is qualitative or contested, that is stated plainly.
The adoption jump is a 39-point rise in a single year: consumers asking AI for local business recommendations rose from 6% to 45% in one year (BrightLocal 2026, panel of 1,002 US adults). Within that panel, 31% had asked ChatGPT for a business recommendation in the past 12 months, and 23% had used Google's AI Mode.
Adoption is broad, not fringe. Among consumers aged 30–44 — prime household-spending years — 64% have asked an AI tool for a business recommendation, against 24% of those over 60 (BrightLocal 2026). The audience behind these percentages is large: OpenAI announced in February 2026 that ChatGPT had reached 900 million weekly active users, up from 800 million in October 2025 (OpenAI, reported by TechCrunch).
At the same time, the click itself is disappearing as the default outcome of a search. Roughly 60% of searches now end without a click to any website, per Bain & Company. A business can hold a strong traditional ranking and still never be seen by the growing share of buyers whose question is answered — and whose shortlist is formed — on the results surface itself.
| Indicator | Figure | Source |
|---|---|---|
| Consumers asking AI for local business recommendations | 45% (vs 6% one year earlier) | BrightLocal, Local Consumer Review Survey 2026 |
| Asked ChatGPT for a business recommendation, past 12 months | 31% | BrightLocal 2026 |
| Used Google AI Mode for recommendations | 23% | BrightLocal 2026 |
| Ages 30–44 who have asked AI for a recommendation | 64% | BrightLocal 2026 |
| Ages 60+ who have asked AI for a recommendation | 24% | BrightLocal 2026 |
| ChatGPT weekly active users | 900 million | OpenAI announcement, Feb 2026 (TechCrunch) |
| Searches ending without a click | ~60% | Bain & Company |
The structural change is the form of the answer. A traditional results page presents many links and leaves the sorting to the buyer; an AI assistant returns a short, synthesized answer that names a small set of specific businesses and explains why. The shortlist is assembled before the buyer visits any website. Businesses absent from that answer are not ranked lower — they are not present at all.
Business buyers show the same pattern with higher stakes. In G2's 2026 buyer research, 51% of B2B buyers now start research with AI chatbots more than with Google, 69% have switched vendors based on AI guidance, and 41% regularly use Deep Research tools that read and summarize dozens of sources before a human ever opens a tab (G2 2026).
The practical consequence is that the moment of consideration has moved upstream. In the link era, a business could be discovered late — page two, a second search, a referral. In the answer era, the recommendation set is composed in one pass from what the AI system can find, verify, and corroborate at that moment.
Fewer clicks does not mean less commercial value per click — the opposite is documented. Adobe's analysis of traffic to US retail and services sites found that visitors arriving from AI sources convert 54% better than visitors from other channels (Adobe). The AI answer performs the comparison work in advance, so the visitor who does click arrives closer to a decision.
Read together, the Bain and Adobe findings describe one funnel replacing another: roughly 60% of searches end without a click (Bain & Company), yet the smaller stream of AI-referred visitors converts 54% better (Adobe). Raw traffic becomes a weaker health metric; presence in the answer, and the quality of the visitors it sends, become the meaningful ones.
The most-cited controlled study of generative engine visibility found that specific, checkable content attributes measurably raise inclusion in AI answers: adding direct quotations increased visibility by 41%, adding statistics by 33%, and adding cited sources by 28% (Aggarwal et al., KDD 2024). The common thread is verifiability — content an AI system can attribute and defend.
Freshness carries independent weight. Content less than three months old averaged 6 AI citations, versus 3.9 for content more than two years old (SE Ranking, 2025). A page that was accurate in 2023 but untouched since competes at a measurable disadvantage against a current one, even when the underlying facts have not changed.
Corroboration matters because AI assistants assemble answers from multiple independent pages — reviews, comparison articles, directories, and the business's own site — rather than trusting any single source. Consistent facts across those surfaces (services, service area, credentials, reviews) give a system more grounds to name a business with confidence; contradictions give it grounds to omit one.
Not every tactic marketed for AI visibility has evidence behind it. In an Ahrefs study of 137,000 domains, 97% of llms.txt files received zero AI-bot traffic (Ahrefs), indicating that file alone does not currently drive AI visibility. The measured levers remain content-side: verifiable claims, current pages, and corroborating third-party coverage.
The measured view: adoption moved from 6% to 45% in a single year (BrightLocal 2026), and even if that pace halves, AI recommendation is on track to be a routine consumer behavior — not a novelty — within the next 12 months. Nothing in the data suggests traditional channels vanish in that window; the evidence describes an added decision layer, not a replaced one.
For a local operator, the evidence supports four low-regret actions over the next year. First, establish a baseline: check what AI assistants currently say when asked for your service in your area, and recheck on a schedule, since answers change. Second, make core pages verifiable in the way the KDD 2024 research rewards — plain claims, real numbers, quotable statements, named sources. Third, keep pages current; the SE Ranking freshness gap (6 vs 3.9 citations) accrues to businesses that update rather than republish. Fourth, tend the third-party record — reviews and independent mentions — because corroboration across sources is what lets a system name you with confidence.
What the evidence does not support is panic spending. The documented levers are content quality, recency, and corroboration — work most operators can schedule — and the buyers arriving through this channel convert 54% better when they do click (Adobe). The rational posture for the next 12 months is measurement plus steady publishing, not wholesale reinvention.
How many people actually use AI to choose businesses? In BrightLocal's 2026 survey of 1,002 US adults, 45% had asked an AI tool for a local business recommendation, up from 6% a year earlier; 31% had used ChatGPT specifically. ChatGPT's overall base reached 900 million weekly active users in February 2026 (OpenAI, via TechCrunch).
Does AI-driven discovery actually produce customers? The available conversion evidence says yes: visitors referred by AI sources convert 54% better than other traffic (Adobe). Volume is lower — about 60% of searches end without any click (Bain & Company) — but the visitors who arrive have already been pre-qualified by the answer itself.
What is the first step for a business that has never checked its AI visibility? Establish a baseline: ask the major assistants for your service in your area and record whether you appear, then apply the evidenced levers — quotations, statistics, and cited sources (+41%/+33%/+28% visibility, Aggarwal et al., KDD 2024), pages under three months old (SE Ranking 2025), and a consistent third-party record.
ASN Research is published by ASN Intelligence LLC. Business owners can run a free AI visibility scan at asncheck.com.