AI Visibility for Personal Injury Lawyers: Prompts, Facts and Access

Injury clients ask AI assistants for help in the hours after an accident, often on a phone, and almost always with a place in the question. These are hypothetical but realistic examples:

• "I got rear-ended on I-35 in Austin yesterday. Do I need a lawyer?"
• "Best car accident lawyer near me who speaks Spanish"
• "How much does a personal injury lawyer take from a settlement in Texas?"
• "Truck accident attorney in Phoenix with free consultation"
• "Does [firm name] handle motorcycle accidents?"
• "Slip and fall lawyer open on weekends in Columbus"
• "Is [firm name] or [other firm] better for a dog bite case?"

Every one of these carries a city, a case type or a fee question. Those are exactly the facts an assistant has to get right to recommend you. OpenAI says ChatGPT search typically rewrites a question into one or more targeted queries, so the exact wording a client types is not what gets searched. Tracking a fixed prompt list is a sample, not a view of real retrieval.

Next move: write ten prompts your own intake team hears every week, in your clients' words. Those become your test set for the rest of this page. The broader guide for law firms covers practice areas beyond injury.

Each engine reaches your site through a different door. Treat them separately.

Google states that a page must be indexed and eligible to show with a snippet to appear as a supporting link in AI Overviews or AI Mode, with no additional technical requirements. There is no separate GEO path into Google.

As of October 2026, retrieval is discretionary. ChatGPT decides per question whether to search, and Claude.ai search is a user toggle. When no search runs, the answer comes from training data with a fixed cutoff. OpenAI lists August 31, 2025 for the Instant model ChatGPT uses by default. A firm that opened an office after that date can be missing from those answers no matter what it does. The free search on vs off test shows how far the two answers about your firm differ, and API answers vs consumer app answers explains why a tool's result may not match what a client sees.

AI platforms get specific facts wrong about specific businesses. For an injury firm, these are the ones to audit first:

• Fee structure. The contingency percentage, whether it changes if a case goes to trial, and who pays case costs if you lose. "No fee unless we win" is often repeated without the cost terms.
• Case types. Whether you take medical malpractice, workers' compensation, wrongful death or trucking cases. An assistant may assume every injury firm handles everything.
• Service area. Which states your lawyers are licensed in, which offices exist, and whether you take cases outside your metro.
• People and offices. Lawyers who have left, old addresses, a former firm name after a merger.
• Practical details. Languages spoken, weekend or after-hours intake, and whether consultations are free.

These errors usually trace to your own site disagreeing with itself, or to old third-party profiles. A fee stated one way on the homepage and another way on a practice page gives an engine two answers to choose from.

The only positive experimental signal for on-page content is a specific fact stated plainly. Put each fact in one clear sentence on a page crawlers can fetch, for example: "We charge 33% of the recovery if your case settles before a lawsuit is filed." (That figure is a hypothetical, not a recommendation.) That gives an engine a sentence it can quote. It does not guarantee it will.

When engines cite other sites instead of yours, being named accurately on those sites is the route into the answer. The evidence shows an association, not a promised effect.

For injury firms, the places worth checking are:

1. Legal directories and lawyer-rating profiles.
2. Your state bar's public lawyer listing.
3. Review platforms and map listings.
4. Local news coverage of verdicts or settlements.
5. Local community threads, including discussion on Reddit.

Open each profile and compare it with your site: fee terms, case types, offices, attorneys, phone number. Fix or claim anything stale. An old directory page that still lists a departed partner can outlive your own correction. More on how much listings matter is in whether industry directories help AI answers.

One risk to know about: some third-party pages carry injected instructions aimed at AI agents. Research on LLM search agents (arXiv:2606.16821) shows manipulated content can change what an agent endorses once it is in the results. Never use the tactic yourself. It is illegitimate, and engine safety updates change how well it works without warning.

Personal injury is crowded, and the tempting shortcut is a page for every suburb: "car accident lawyer in [town]" times two hundred. Google's spam policy defines scaled content abuse as many pages "generated for the primary purpose of manipulating search rankings and not helping users," and names generative AI as its first example.

Otterly tested volume publishing in 2026 with two new domains of roughly 1,000 AI-generated posts each. Otterly reports Google deindexed one site after about 14 days and the other after about 90, with zero manual actions in Search Console on either. Once Google dropped them, Google's AI features almost never cited them. Copilot kept citing them because Bing kept them indexed.

The test was one pair of low-stakes sites, so it proves less than its headline. The policy lesson still holds. Google says appropriate use of AI is fine. The violation is volume without value. For an injury firm, losing Google takes AI Overviews and AI Mode with it.

Next move: keep a page for each real office, and write local pages only where you have something specific to say, such as the courthouse, local filing deadlines or cases you actually handled there.

The levers with real evidence behind them are access levers. They work by removing blocks.

robots.txt, bot by bot

robots.txt rules are per bot. OpenAI states each setting is independent: GPTBot governs training, and OAI-SearchBot governs ChatGPT search. A blanket AI block often removes search eligibility by accident. A setup that allows search bots while opting out of OpenAI training looks like this:

User-agent: OAI-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: GPTBot
Disallow: /

Firewall and CDN

A firewall or CDN can block a crawler that robots.txt allows, often with a 403. OpenAI says eligibility needs both the robots.txt permission and a host or CDN that allows its published search bot IP addresses. From September 15, 2026, Cloudflare blocks Training and Agent crawlers by default on ad-displaying pages for new domains and existing free-tier customers, while Search stays allowed. Check your settings rather than assume. The AI Crawler Checker requests a page as each AI crawler and shows which ones your firewall turns away.

Rendering and snippet controls

Vercel found ChatGPT's and Claude's crawlers fetch JavaScript but don't execute it. If your fee details or case types load in a script widget, those crawlers may never see them. Also search your templates for noindex, nosnippet, max-snippet:0 and data-nosnippet, which keep content out of AI answers. Finally, check the generative AI control in Search Console, where available. Inclusion is the default, but child properties inherit the parent's setting. The ChatGPT citation guide covers OpenAI's side in more depth.

Say you run a hypothetical four-lawyer injury firm with two offices. A month of work looks like this.

1. Week 1: access. Read robots.txt for each search bot. Test each crawler against your firewall. Confirm fee and practice-area text appears in the raw HTML before JavaScript runs. Remove any accidental snippet blocks.
2. Week 2: your own facts. List fee terms, case types, licensed states, offices, attorneys, languages and intake hours. Find every page that states them and make them agree. Write each one as a single plain sentence.
3. Week 3: third parties. Audit your bar listing, directory profiles, review sites and map listings against that list. Fix what you control and request corrections for the rest.
4. Week 4: test and record. Check index status in Google Search Console and Bing Webmaster Tools separately. Run your ten prompts on each engine, with search on and off where you can, and note every wrong fact. The free checker on this site gives a snapshot for up to three buyer questions.

Repeat week 4 monthly. Model releases and index changes move answers without any change on your side.

Will schema markup or an llms.txt file get my firm cited? No evidence says so. Google states structured data isn't required for its generative AI features, and Ahrefs' matched test found AI Overview citations fell 4.6% after pages added schema. Google Search ignores llms.txt, and no other engine documents reading it. Existing markup does no harm, so there is no need to remove it. Spend the time on crawler access and consistent facts instead.

If I block GPTBot, do I disappear from ChatGPT? No. GPTBot controls whether your content may be used to train OpenAI's models. ChatGPT search uses OAI-SearchBot, and OpenAI states each setting is independent of the others. You can block GPTBot and still allow OAI-SearchBot. The reverse also holds: allowing GPTBot does nothing for ChatGPT search if OAI-SearchBot is blocked by robots.txt or by your firewall.

Do online reviews decide which injury lawyer AI recommends? Nothing published shows that. One widely shared study on review profiles never defined its citation measure and cannot separate reviews from firm size. What holds is narrower: review sites and directories are places engines may cite, so the facts on them should be accurate. Fix wrong fee terms, practice areas and offices there. Never post fake reviews or pay for them.

Will a new AI model change what it says about my firm? It can. Answers without search come from training data frozen at a cutoff, and a new model brings a new cutoff and new behaviour. Answers with search depend on what the engine retrieves that day. That is why a one-time check goes stale. Read [how new model releases change brand answers](/questions/new-ai-model-changes-brand-answers) and retest your prompts after major launches.

Does a strong Google ranking mean AI Overviews will mention my firm? Not necessarily. Ahrefs found that 37.9% of URLs cited in AI Overviews also appeared in the first 10 results in March 2026, so most citations came from further down. Google also says indexing and serving isn't guaranteed. Rankings still matter for clients who click, but treat AI answers as a separate result to check. [AI search vs Google search](/guides/ai-search-vs-google-search) explains the differences.