How US Law Firms Get Found on ChatGPT, Google AI & AI Search

Key Takeaways:  A major share of AI visibility for US law firms is shaped outside your website, especially when AI systems are deciding whether to recommend your firm rather than cite a page you wrote. To get found across ChatGPT, Google AI Overviews, Gemini, Perplexity, Claude, and Microsoft Copilot, firms need three things working together: clean on-site structure, strong off-site authority, and a compliance workflow aligned with ABA Formal Opinion 512 (July 29, 2024) and your state's Rules of Professional Conduct.

Want to see where your firm stands? SkyScale can benchmark your firm against competitors across ChatGPT, Google AI Overviews, Perplexity, Gemini, Claude, and Copilot.

The thing most law firm AI guides skip

Read three of the agency posts currently ranking for this query and you'll come away thinking the answer is schema. Add Attorney schema. Add FAQPage. Tag your practice areas with LegalService. Ship.

That's not wrong. It's just the smallest piece of the picture. A major share of whether ChatGPT names your firm, or whether Google's AI Overview pulls from your page, is shaped outside your website, especially when the AI is deciding whether to recommend your firm.

ChatGPT runs in two modes. In its default mode, it answers from training data sourced largely from Common Crawl and other large public corpora. Your firm appears there only if it left a fingerprint big enough to survive. In ChatGPT Search, OpenAI describes the product as combining web search results with the model's reasoning, returning answers with source links, drawing on live retrieval and search partnerships, which include Bing-indexed results in many cases.

What this means in practice: when somebody asks ChatGPT "best personal injury lawyer near me" or "top truck accident attorney in Texas," the names that surface are heavily shaped by what already lives on directories, legal press, Wikipedia, Reddit, and high-citation legal databases. Your site is one input. The internet's accumulated mention of your firm is the bigger one.

Two different games: cited as a source vs recommended as a firm

There are two distinct wins in AI search, and they require different work.

Cited as a source. A potential client asks an informational question, "what's the statute of limitations on a slip-and-fall in Pennsylvania," "how does parental relocation work in Florida custody cases", and the AI pulls from one or more attorney-authored pages to build its answer. This is where content quality, schema, and answer-first structure pay off, the discipline sometimes called Answer Engine Optimization. A solid practice-area page can win citations almost mechanically if it's structured right.

Recommended as a firm. A different query, "best litigation lawyer in New York," "who handles medical malpractice lawsuits," "top rated injury law firms," "best estate litigation lawyer nearby", pulls from a different set of signals. The AI is making a quasi-recommendation and the models are conservative. They lean on accumulated brand mentions, directory presence, reviews, press, and consistent firm-entity signals across the web. Flawless schema won't get you here if your firm isn't a recognized entity outside its own domain.

Most firms apply the same playbook to both. The playbook for source citation is editorial. The playbook for firm recommendation is reputational. They are not the same project.

How each platform picks law firms

ChatGPT Search. Combines live web retrieval with the model's reasoning, returning answers with source links. Pages with clear claims, specific statistics, primary-source citations, and direct answers near the top tend to be easier for the system to retrieve and quote. Question-style H2s with the answer in the first 40–60 words do well. A focused ChatGPT SEO strategy typically lives at the intersection of Bing-index quality, answer-first structure, and recency signals.

ChatGPT (default, no retrieval). Older training cuts. Your firm appears here if you've accumulated mentions, directory listings, press, and citations that survived in the training corpus. You can't influence this in the short term.

Google AI Overviews. Built on Gemini. Google's own documentation is explicit that AI features are rooted in core Search ranking and quality systems, the same things that already make pages rank in Google also drive AI Overview inclusion. Freshness helps when a topic genuinely changes, but it isn't a separate freshness-only system.

Gemini (standalone). Closely integrated with Google's Knowledge Graph, Google Business Profile, and Maps. For local legal queries, Business Profile completeness and review volume matter more here than on any other platform. Practical work for Gemini visibility leans heavily on Business Profile depth, review velocity, and Knowledge Graph alignment.

Perplexity. Most transparent about its sources. Rewards content structured like a research brief: clear claims, sourced statistics, links to primary law. Perplexity SEO is essentially research-grade citation discipline applied to your own content.

Claude. Doesn't browse by default in consumer chat. Tends to favor longer, nuanced content with full citations. Closest to a careful research assistant. Claude SEO work rewards depth and attribution over snippet-bait.

Microsoft Copilot. Built on Bing's index, integrated into Windows and Microsoft 365. Bing-index optimization helps both Copilot and ChatGPT Search retrieval.

Quick comparison: AI surfaces, signals, priorities

‍

The 45-minute AI visibility audit

Before you change anything, find out where you stand.

Step 1: Firm-recommendation queries. Open ChatGPT, Gemini, Perplexity, and Claude. Run prompts that ask the AI to name a firm:

• "Best personal injury lawyer near me"
• "Top truck accident attorney in [your state]"
• "Best litigation lawyer in [your city]"
• "Top rated injury law firms in [your city]"
• "Law firms experienced with catastrophic injury cases in [your state]"

Run each query twice. AI outputs vary, and the variance tells you whether you're on the edge of the model's confidence threshold.

Step 2: Informational queries.

• "How do I find a trusted attorney?"
• "Who handles medical malpractice lawsuits?"
• "What's the statute of limitations for personal injury in [your state]?"
• "What should I do after a car accident in [state]?"

Note which sources the AI cites. If you only see FindLaw, Nolo, Justia, and .gov pages (no law firm sites), that's the source-citation real estate available in your topic.

Step 3: Brand check. Search your firm name on Perplexity. If citations are mostly your own site, you have a brand-mention problem.

Step 4: Competitor comparison. Run the same queries with a competitor's name. The gap is your work plan.

Step 5: Document. A two-column note: where you appear, where you don't. Repeat quarterly.

‍

On-site foundation

On-site work is the price of admission, the layer most agencies sell as the whole solution when it's really one of three. Done well, AI SEO for law firms treats schema, structure, and attribution as a connected system rather than isolated boxes to tick.

Schema. Six Schema.org types carry most of the load:

• LegalService on practice area pages, areaServed tied to specific cities and counties
• Attorney or Person on every bio, with knowsAbout, alumniOf, and visible bar admission state
• LocalBusiness on office pages with accurate NAP and geo
• FAQPage with real client questions, not marketing questions
• Article on every post, author connected via @id to the attorney's Person schema
• Review and AggregateRating only when legitimate, faking exposes you under ABA Model Rule 7.1

Connect every block via @id so the whole site reads as one entity graph. Validate with Google's Rich Results Test.

Answer-first structure. Lead with the answer in the first 40–60 words. Expand into nuance, exceptions, and jurisdiction afterward. Pages that bury the answer don't get cited.

Question-shaped H2s. "How long do I have to file a personal injury claim in Virginia?" outperforms "Personal Injury Statute of Limitations" because it matches how clients prompt AI tools.

Visible attorney attribution. Every substantive page shows a reviewing attorney's name, bar admission state, and a link to their bio. ChatGPT in particular relies on visible content because it doesn't always parse schema the way Google does.

Last-updated dates. Freshness helps when a topic genuinely changes. Quarterly refreshes with substantive updates (not just touched timestamps) keep content competitive.

Mobile and speed. Standard but non-negotiable.

Crawl access. Most law firms should allow GPTBot, ClaudeBot, PerplexityBot, and Google-Extended in robots.txt. Blocking removes you from those models' training and retrieval.

Content that actually gets cited

The standard is narrower than most agencies imply.

Specific over general. "In California, personal injury claims must be filed within two years under California Code of Civil Procedure §335.1" gets cited. "Statutes of limitations vary by state" does not.

Jurisdiction-named. Say what state in plain English, not in fine print. AI systems are conservative on YMYL content.

Attorney-attributed. Not "Our Legal Team." A named attorney with a bar number and a real bio.

Linked to primary law. Statutes, regulations, court opinions, not other agency blog posts. Most federal sources are at Cornell Legal Information Institute.

Q&A structured. Real client questions in real client language. "Can I sue if I slipped in a grocery store parking lot?" not "Premises liability standard for invitees."

TL;DR blocks at the top of long content. AI extraction loves a labeled summary.

What doesn't get cited: 600-word generic blog posts, pages that delay the answer, marketing-brochure practice areas, anything ghostwritten and unattributed.

Off-site authority is the bigger lever

The training data behind ChatGPT, Claude, Gemini, and other models is dominated by Wikipedia, Reddit, established news, government domains, legal directories, and academic content. Your website matters; the rest of the internet matters more.

The US off-site authority stack:

• Legal directories. Justia, FindLaw, Avvo, Martindale-Hubbell, Super Lawyers, Best Lawyers, Chambers USA, Legal 500. Maintain substantive profiles, not skeletons.
• State and local bar profiles. Free and underused. AI systems trust .org and .gov.
• Local and trade press. Law360, ABA Journal, regional business journals, state legal news sites.
• Wikipedia and similar references. If your firm or any attorney qualifies for a notable mention, high-leverage.
• Reddit, Quora, forums. Not for self-promotion, that violates platform and bar rules. But organic mentions get absorbed.
• Podcasts and video. Long-form audio with transcripts published on the web gets pulled into training and retrieval.
• Authoritative backlinks. Still meaningful.
• Google Business Profile. Foundational for Gemini and local AI Overviews.

Schema gets you to extractable. Off-site presence gets you to known.

‍Case study: how a mid-size PI firm rebuilt AI visibility in two quarters

Composite scenario drawn from real engagement patterns. Firm details anonymized.

A 14-attorney personal injury firm in a tier-2 Texas metro came in after a partner watched ChatGPT recommend three competitors when he tested the prompt "best truck accident attorney in [his city]", and his firm wasn't named once. Organic traffic was steady. Phone volume had quietly dropped 18% over six months. The marketing director assumed it was seasonal.

It wasn't.

The audit (week 1). Forty-five minutes of prompt testing across ChatGPT, Gemini, Perplexity, and Claude returned a clear pattern. The firm appeared inconsistently on informational queries ("how long do I have to file a truck accident claim in Texas") and never on firm-recommendation queries. Citations on informational queries went to FindLaw, Nolo, and the Texas Department of Insurance, almost no law firm sites at all. On Perplexity, searching the firm name returned mostly the firm's own pages back. Brand-mention problem confirmed.

Where they actually stood: strong Google rankings (top 5 for several commercial keywords), thin schema (a single LegalService block on the homepage, nothing else), generic "Our Legal Team" bylines on every blog post, complete Avvo and Martindale profiles but a Justia page abandoned in 2021, no Google Business Profile posts in over a year, and zero podcast or trade-press placements in the prior 24 months.

The 90-day plan (what they actually did).

Days 1–14: Schema rebuilt across attorney bios, practice area pages, and three office locations, all cross-referenced via @id. Justia profile reclaimed and expanded with attorney narratives and case results (within the bounds of Texas Disciplinary Rule 7.02 on advertising). Top three practice area pages rewritten so the first 100 words answered the question implied by the page title. Generic team bylines replaced with named attorney attribution plus bar admission state.

Days 15–60: Real FAQ blocks added to top pages, questions sourced from actual client intake calls, not marketing brainstorms. Five top-performing blog posts refreshed with statute references (Texas Civil Practice and Remedies Code §16.003), recent appellate decisions, and TL;DR blocks. Two attorneys placed on a regional plaintiffs' bar podcast; transcripts published to the firm site and the host's. Google Business Profile reactivated with weekly posts about case updates and community involvement.

Days 61–90: New content cadence, three attorney-authored pieces per month answering questions the audit had revealed AI systems were citing other sources for. A bylined op-ed placed in the regional business journal on commercial trucking liability. State bar profile updated. Local press picked up a community sponsorship the firm had been quietly doing for years.

What changed by week 16.

The same audit, re-run on the same prompts:

• Firm now appeared in 4 of 6 ChatGPT firm-recommendation queries for their primary practice areas in their metro. Still inconsistent on the most competitive ("best truck accident attorney in Texas" statewide), but present locally.
• Cited as a source on 3 informational queries where they had previously been invisible, including "truck accident statute of limitations Texas", the page that had been rewritten in week two.
• Perplexity now returned the firm with third-party citations alongside the firm's own site (the podcast, the op-ed, the updated Justia profile).
• Gemini surfaced the firm on local queries with the refreshed Business Profile.

The honest part. Phone volume recovered roughly two-thirds of the dip by the end of quarter two, with intake forms showing a small but growing share of clients answering "ChatGPT" or "Google AI" to the how did you find us question. The firm hasn't broken into the most competitive statewide AI recommendations yet, that's the work of quarters three through six, and it depends largely on more press and directory authority, not on more on-site work.

What carried the result. Not the schema, though the schema mattered. The biggest movers were the named attorney bylines, the rewritten answer-first openings, the reclaimed Justia profile, and the podcast placement. In other words: a mix of on-site editorial discipline and off-site authority work, applied in the right order.

Practice-area realities

High-value legal searches don't follow one pattern. A mass tort claimant searches around a fast-moving litigation campaign. A truck accident victim needs representation today. A family planning an estate compares local firms carefully before contact. Each practice area behaves differently in AI search, and a generic playbook leaks budget in whichever lane it doesn't fit.

Below: how each major practice area behaves in AI search, and where to focus.

Mass tort. High-value, campaign-driven searches around defective drugs, medical devices, toxic exposure, and consumer claims. AI systems are conservative here for compliance reasons, claimant qualification, filing deadlines, and proof of authority all matter. Firms that win build authority around specific litigations (named drugs, named devices) with attorney-attributed pages, MDL tracking, and clear deadline guidance. State bar advertising and solicitation rules apply aggressively; tread carefully.

Personal injury. The most competitive AI search environment for US law firms. Saturated local markets, AI Overviews trigger frequently. Car accidents, premises liability, negligence, insurance disputes. What pays off: deep practice-area pages by injury type, comprehensive FAQs on insurance and settlement timing, reviews, location relevance, and aggressive third-party citation. Most crowded share-of-voice landscape.

Truck accident. Urgent intent. Commercial vehicles, FMCSA issues, severe injuries, complex liability against corporate defendants. AI systems weight specific regulatory citation (FMCSA hours-of-service rules, federal trucking regulations) and named-attorney credibility. Local intent dominates, clients want a lawyer who handles trucking cases in their state.

Medical malpractice. Trust-heavy searches around surgical errors, delayed diagnosis, birth injuries, hospital negligence. AI systems are notably cautious because of stakes. Named attorney credentials, board certifications, expert witness affiliations, and case experience move the needle more than in less-sensitive areas. Standards-of-care content with primary-source citation outperforms generic explainers.

Catastrophic injury. High-value searches involving traumatic brain injuries, spinal cord injuries, amputations, burns, and permanent disability. AI systems lean toward firms with documented trial experience, life-care planning depth, and long-term damages calculations. Pages that go shallow on damages math get skipped.

DUI. Urgent local intent. License suspension, breath tests, field sobriety, court dates, penalties, prior offenses. Clients want an attorney they can contact today. AI Overviews trigger on procedural questions ("what happens at a DUI arraignment in [state]") and Gemini surfaces local firms with strong Business Profile signals.

Criminal defense. Urgent intent across arrests, charges, investigations, court appearances, bail. AI systems are conservative due to stakes. Local defense experience, attorney availability, and visible trust signals (named attorneys, bar admissions, prior verdicts within bar-rule limits) outperform generic content. Same-day responsiveness signals matter.

Family law. Emotionally sensitive searches around divorce, custody, support, asset division, and protective orders. Strong informational query volume. Jurisdiction-specific content has the highest leverage because family law is state-defined and AI is cautious about misapplying state rules. Local court experience and reviews carry weight. Less competition than PI in many markets.

Estate planning. Trust-led local searches around wills, trusts, powers of attorney, asset protection, succession planning, and elder planning. Lots of informational long-tail. AI Overviews trigger consistently. Tax-sensitive content and state-specific probate detail win disproportionately.

Immigration. Question-driven searches around visas, green cards, deportation defense, asylum, family petitions, and work authorization. Highly federal. AI systems default to USCIS and a handful of well-known firms and nonprofits (AILA, American Immigration Council). Multilingual content demand is real. Policy changes drive freshness requirements. Hard to break into the cited set, sticky once you're in.

Employment. Split intent, employees searching for "can I sue my employer" and employers searching for "non-compete enforceability." Wrongful termination, discrimination, harassment, retaliation, wage disputes, severance. Filing deadlines and proof-of-authority requirements vary heavily by state and by EEOC vs state-agency tracks. Signal clearly which side you serve.

Bankruptcy. Distressed financial searches around Chapter 7, Chapter 13, creditor harassment, foreclosure, wage garnishment, and automatic stays. Strong informational query environment around debt timelines and procedural questions. AI Overviews trigger frequently. Local intent dominates because filings are jurisdiction-specific. Clear next-step content with named attorney attribution outperforms generic debt-relief copy.

General rule. Consumer practice areas reward on-site content depth and EEAT. Commercial and high-stakes practice areas reward off-site authority, directories, press, expert credentials, and trial track record. Most firms run a single playbook across all their practice areas and leak ground in whichever lane it doesn't fit.

Compliance: the part that actually puts your license at risk

ABA Model Rules aren't binding law. Each US state adopts its own version of the Rules of Professional Conduct, and those, along with state ethics opinions, govern your practice. Treat the Model Rules and ABA opinions as persuasive starting points; your state's rules are the law.

Model Rule 7.1 prohibits false or misleading communications about a lawyer or their services. AI tools can produce content that crosses this line easily. If you publish AI-drafted content under an attorney's name, the attorney owns everything in it.

Rules 7.2 and 7.3 govern advertising and solicitation. AI-generated outreach, mass-personalized email, and chatbots that initiate contact with potential clients sit in or near this zone.

ABA Formal Opinion 512 (July 29, 2024) is now the de facto baseline. It addresses competence (1.1), confidentiality (1.6), client communication (1.4), supervision (5.1, 5.3), candor to the tribunal (3.3), and reasonable fees (1.5). The central message: lawyers must understand the AI tools they use, supervise outputs, and verify what those tools produce. Publishing an AI-drafted blog post under an attorney byline with no real review is a supervisory failure.

State-level variation. California, Florida, New Jersey, New York, and Pennsylvania have all issued their own GAI ethics guidance. Some require disclosure to clients. Some restrict automated communications. Check your state's most recent opinions before deploying anything client-facing.

Sanctions environment. Federal courts have sanctioned attorneys for fabricated AI citations dozens of times since 2023, including a Sixth Circuit sanction in early 2026 for a brief with more than two dozen hallucinated citations.

Practical rules:

1. Use AI for drafting and structure, not for legal substance.
2. Every published page lists a named reviewing attorney with a review date.
3. Maintain a documented review workflow. "An AI wrote it" is not a defense.
4. Don't fake reviews, awards, or ratings.
5. Disclose AI use where your state requires it.

AI changes what you can produce, not what you're responsible for.

Tracking and measuring

Free baseline.

• The 45-minute audit, quarterly
• Google Search Console: feeds AI Overview signal indirectly
• Google Business Profile insights
• Manual prompt tests across ChatGPT, Gemini, Perplexity, Claude, logged in a spreadsheet
• GA4 custom channel grouping for AI referrals: chat.openai.com, perplexity.ai, gemini.google.com, copilot.microsoft.com, claude.ai

Paid (scale-dependent). AI citation trackers, Profound, Otterly, Peec.AI, AthenaHQ. Useful at scale, overkill for solos. Ahrefs Brand Radar and Semrush are adding LLM modules.

Track: queries where you appear monthly, share of voice vs 3–4 named competitors, AI-referral traffic, branded search volume, intake-form "how did you find us" with AI option.

Ignore: daily citation fluctuations (AI is non-deterministic), pure schema scores, "guaranteed citation" pitches. No vendor can guarantee that.

A realistic 90-day plan

Days 1–14: audit and triage. Run the 45-minute audit. Audit schema with Rich Results Test on homepage, two practice-area pages, two attorney bios. Pull top 20 ranking pages from Search Console. Verify all directory profiles are claimed.

Days 15–30: foundation. Rebuild attorney bio schema first, then practice area pages, then office pages, all connected via @id. Rewrite the first 100 words of your top three practice-area pages to lead with the answer. Add real FAQ blocks. Add bar admission state to every byline.

Days 31–60: content depth. Refresh five top-performing posts: update stats, add a 2026 case reference, add TL;DR, update last-updated date. Publish three new attorney-reviewed pieces per month built around real client questions. Add primary-source citations everywhere.

Days 61–90: off-site + measurement. Place two attorneys in podcasts or trade publications. Pursue two new directory listings. Set up GA4 AI-referral tracking. Re-run the audit.

Quarter four is the same shape on repeat.

AI visibility checklist

On-site

•   Attorney / Person schema on every bio
•   LegalService schema on every practice area
•   LocalBusiness schema on every office
•   FAQPage schema with real client questions
•   Article schema on every post with @id author link
•   All schema cross-referenced through @id
•   Visible attorney byline + bar admission state
•   Answer-first opening paragraphs on top pages
•   Question-shaped H2s
•   TL;DR blocks on long-form
•   Mobile speed and Core Web Vitals reviewed

Content

•   Attorney-reviewed FAQs on every practice area
•   Jurisdiction named explicitly
•   Primary-source links (statutes, regulations, opinions)
•   Quarterly substantive refresh

Off-site

•   Directory profiles complete and consistent (Justia, FindLaw, Avvo, Martindale, Super Lawyers)
•   State bar profile claimed
•   Google Business Profile complete and posting
•   NAP consistent everywhere
•   One podcast or trade-press placement per quarter

Measurement

•   GA4 AI referral channel grouping
•   Quarterly prompt audit logged
•   Share-of-voice tracking vs 3–4 competitors
•   Intake form: "How did you find us?" with AI option

Compliance

•   Named attorney reviewer for every page
•   Documented review workflow
•   State Rules of Professional Conduct checked for AI disclosure
•   No falsified reviews, awards, or ratings
•   ABA Formal Opinion 512 read by AI supervisor

The future of legal SEO is recommendation visibility

Traditional rankings still matter. But legal search is shifting toward recommendation-driven discovery.

The firms that win over the next few years won't simply rank. They'll become trusted entities, authoritative sources, locally reinforced brands, and repeatedly cited legal experts inside AI-generated answers. That requires more than publishing blogs. It requires building a complete trust ecosystem AI systems can confidently interpret and recommend.

For many law firms, the biggest mistake right now is assuming AI visibility is just another SEO trend. It isn't. It's a behavioral shift. Potential clients are already asking:

• "Who is the best personal injury lawyer near me?"
• "Which law firm handles serious truck accident cases?"
• "What attorney is trusted for medical malpractice claims?"

…inside ChatGPT, Google AI Overviews, Perplexity, and Gemini before they ever browse traditional search listings. The firms appearing there are shaping trust earlier in the decision-making journey. That advantage compounds.

The law firms that adapt early will build stronger authority, better recommendation visibility, higher-intent consultations, greater brand trust, and long-term competitive insulation, while slower firms risk becoming increasingly invisible in AI-first discovery.

At SkyScale, the focus isn't just ranking websites. It's helping law firms become discoverable, understandable, and recommendable across modern AI search systems, through AI SEO, Generative Engine Optimization (GEO), entity SEO, topical authority, structured data, and conversational search optimization. The full set of AI search services maps to the platforms that actually matter: ChatGPT, Gemini, Perplexity, and Claude.

Because the next phase of legal search won't be won by whoever publishes the most content. It'll be won by the firms AI systems trust enough to recommend.