AI Search Trust Just Dropped 28 Points. Your Firm's Content Needs to Compensate.

A new consumer survey landed in mid-June, and it should change how you think about the next AI Overview that names your firm. Fractl and Search Engine Land polled 1,008 U.S. consumers and found that the share who rate AI search as more helpful than traditional search fell from 82% to 54% in a single year — a 28-point collapse. Over the same period, the population of AI skeptics grew sixfold, from 3% to 17%. And yet 70% of those same consumers say they're using AI search more than they did a year ago. Only 3% pulled back.

That's not a contradiction. It's habituation. People have folded AI search into how they look things up — it's fast, it's convenient, it's the path of least resistance — while growing steadily more suspicious of what it tells them. For a firm that's started showing up in ChatGPT answers and Google AI Overviews for "H-1B attorney near me" or "what to do after an RFE," this matters more than almost any other stat in your marketing stack. Being named by the AI is no longer the finish line. It's the starting gun for a skeptical reader deciding whether to believe you.

The skepticism tax

Immigration searches carry more urgency than most. Someone typing "priority date retrogression EB-2" or "can I still apply for adjustment of status after a denial" is usually under real time pressure, sometimes real fear. If an AI assistant surfaces your firm as an option, that searcher isn't handing you their trust — they're handing you a shortened list of finalists to interrogate. The study found the share who say heavy AI use by a brand makes them trust that brand less nearly doubled, from 20% to 39% year over year, and hits 54% among Gen Z. Your prospective client has learned, correctly, that AI-generated summaries can be wrong, generic, or quietly promotional. They're looking for a reason to doubt the answer before they book a consultation off the back of it.

That means the content sitting behind the citation is doing more work than it used to. A practice-area page that reads like every other firm's practice-area page — "years of experience," "dedicated to our clients," no names, no numbers — doesn't survive that skepticism tax. It confirms it.

What survives skepticism

Two findings from the same research cycle point to what does hold up. Ahrefs analyzed 75,000 brands across ChatGPT, AI Mode, and AI Overviews and found that branded web mentions — people and publications naming your firm by name — correlate with AI visibility at 0.664 to 0.709. Backlinks, the old SEO currency, correlate below 0.30. Separately, Fractl found that 84–91% of consumers want AI-assisted content clearly labeled, but only 20% of brands consistently disclose it. The gap between what readers want and what brands provide is exactly where trust leaks out.

Translate both into what a small immigration practice controls directly: named attorneys, not "our team." Bar admission numbers, not vague credentials. Dated outcomes — "RFE overcome, EB-1A petition, filed March 2026" — not "successful track record." A citation is only as trustworthy as the page it points to, and right now most firm websites are giving a skeptical reader nothing to verify.

Verifiable beats polished

This is the part that's counterintuitive if you've been thinking about your website the way a SaaS company thinks about its landing page. Polish doesn't rebuild trust that's declining industry-wide — specificity does. A page that names the attorney who handled a naturalization case, states the filing date, and links to the underlying USCIS policy memo reads as harder to fabricate than a beautifully designed page making the same claim in the abstract. Skeptical readers are pattern-matching for exactly that kind of texture. It's also, not coincidentally, the kind of content AI systems weight more heavily when deciding what to cite.

Do this before Friday

You don't need a new content strategy to respond to this. You need to audit what's already published and close the gaps that make it easy to distrust.

1. Add a named attorney byline to every practice-area page — full name, bar number, one line on relevant case types (H-1B, asylum, removal defense, consular processing). No "our team" pages standing alone.
2. Date every case result or outcome you publish, even briefly. "Filed [month/year]" or "resolved [month/year]" turns a vague claim into a verifiable one.
3. Replace one abstract claim per page with a specific number — cases handled this year, average RFE response time, consultations completed. Precision reads as evidence; adjectives read as marketing.
4. Search your own firm name in ChatGPT and Google's AI Overview. If either misdescribes your practice areas, your location, or a case outcome, that's a correction worth making on your site now, since inaccurate AI descriptions are hard to unwind once they've been cited elsewhere.
5. Add a one-line disclosure note anywhere you use AI-assisted drafting for blog content or FAQs. It costs you nothing and puts you ahead of the 80% of brands that don't bother.

None of this requires a redesign. It requires treating every page a prospective client might land on after an AI citation as evidence, not advertising.

The bigger picture

Trust in AI search isn't going to recover on its own — the study's authors are blunt about that. But the firms that get named and believed won't be the ones with the flashiest content. They'll be the ones whose content survives a skeptical second look: real attorneys, real dates, real numbers, nothing left to take on faith. That's a smaller bar than most firms think, and most competitors haven't cleared it yet.

Clientory helps immigration law firms become visible to the families and individuals who need them most — turning AI search into consultations, and consultations into clients.

See where your firm currently stands in AI search results — book a free visibility check.