Law Firms Have a New Audience: AI

As automated traffic grows and consumers turn to AI for legal questions, law firms need to make their expertise visible to machines as well as people.

A prospective client has just been injured. Before contacting an attorney, she asks an AI assistant what her options are, how the legal process works, and which firms handle cases like hers. By the time she visits a law firm's website, she may already have a shortlist.

For decades, law firm marketing has focused on getting a person to find a website, read it, and make contact. That process is changing. AI systems are increasingly helping people interpret information and evaluate potential legal representation before a website visit ever occurs.

My view is that making a law firm's expertise visible to these systems is becoming a mission-critical part of marketing. The evidence does not suggest that AI is replacing lawyers or that AI crawlers have overtaken human visitors. It does suggest that the audience evaluating a firm's public information is changing.

Bots already generate more than half of web traffic in one major dataset

In its 2026 Bad Bot Report, cybersecurity company Thales estimated that bots accounted for 53% of global web traffic in 2025, compared with 47% attributed to humans.

Figure 1

Bots vs. humans: share of web traffic in two datasets

  • Automated (bots and AI agents)
  • Human

Thales · global web traffic, 2025

DataDome · its customers' sites, July 2025 to June 2026

Thales counts all bots, including search crawlers and malicious automation. DataDome calls its figure a floor because its sites sit behind upstream bot filtering. The two measure different networks and are not directly comparable. Sources: Thales, 2026 Bad Bot Report; DataDome, State of Bot & Agent Security Report 2026.

The distinction matters. Those 53 percentage points include ordinary search crawlers, security tools, malicious automation, and other bots. They are not all AI assistants researching businesses for prospective clients. DataDome, whose traffic sample often sits behind upstream bot filtering, found a substantially lower automated share of approximately 26.5%. Bot-traffic percentages are highly dependent on the network and measurement method.

A more focused measure comes from Cloudflare, which found that identified AI bots generated an average of 4.2% of HTML requests in 2025, with dual-purpose Googlebot measured separately. DataDome reported that AI traffic represented 1.7% of all requests it observed by June 2026. These figures measure different networks and types of requests and should not be treated as directly comparable.

The significance is not that AI already accounts for most website traffic. It is that a distinct source of automated information gathering is growing rapidly.

AI traffic is growing much faster than human traffic

Across the sites DataDome analyzed, AI-generated requests increased 82.3% from July 2025 to June 2026. Human requests grew 13.2% over the same period, according to its 2026 State of Bot & Agent Security Report.

Figure 2

Growth in requests, July 2025 to June 2026

  • AI agents and crawlers
  • Humans

AI agent and crawler requests

Human requests

Change in request volume across the sites DataDome protects. Growth rates, not shares of traffic: AI was 1.7% of all requests DataDome observed by June 2026. Source: DataDome, State of Bot & Agent Security Report 2026.

A separate analysis by HUMAN Security reported that monthly AI-driven traffic rose 187% between January and December 2025. Its research also found that much of this activity was still related to large-scale information collection rather than direct customer interactions.

These findings establish growth, not commercial value. A crawler request is not a prospective client. But they show why firms should pay attention to how their information is accessed and interpreted by automated systems.

The evidence is especially relevant to law firms

The most important statistic for the legal industry may have little to do with servers.

In its 2026 Legal Trends Report, Clio found that 24% of U.S. adults had already used AI to answer a legal question, while 46% said they would consult AI before contacting a lawyer. These are separate measures of actual behavior and stated willingness, not a claim that 46% have already used AI to find a law firm.

Figure 3

U.S. adults and AI for legal questions

Have used AI to answer a legal question

Would consult AI before contacting a lawyer

Two separate measures: what people have done, and what they say they would do. Neither is the share who have used AI to find a law firm. Source: Clio, 2026 Legal Trends Report.

This matters because the early stages of choosing an attorney often involve sensitive, complicated questions. Someone facing an employment dispute, a serious injury, or a family crisis may want to understand the relevant issues before speaking with a professional. An AI assistant can organize that initial research, although it cannot replace qualified legal advice.

The wider search market offers another signal. A Pew Research Center study of Google searches found that users clicked a traditional search result in 8% of visits when an AI summary appeared, versus 15% when no AI summary appeared.

The finding is not specific to legal services and does not establish causation. It does show that a website can contribute to a person's research without receiving the click that marketers traditionally rely on to measure exposure.

For law firms, this creates a practical problem: a firm may have substantial experience and a strong referral reputation, yet be poorly represented, omitted, or inaccurately described when someone asks an AI system about relevant counsel.

What marketing to AI systems actually means

Marketing to bots does not mean persuading malicious crawlers, publishing hundreds of generic AI-written articles, or abandoning conventional search optimization.

It means ensuring that legitimate search systems and AI assistants can find accurate, useful, verifiable information about the firm.

That work begins with four priorities.

1. Make important information accessible.

Check whether the firm's website, hosting configuration, and security rules allow legitimate crawlers to read public practice and attorney pages. OpenAI, for example, distinguishes OAI-SearchBot, which supports ChatGPT search, from GPTBot, which is used for potential model training.

Firms can make deliberate choices about access rather than treating all bots alike.

2. Publish substance that reflects real expertise.

A practice page should explain what matters in a case, which jurisdictions the firm serves, and how its attorneys approach the work. Attorney biographies should identify demonstrable experience. Public case descriptions must be accurate and respect client confidentiality.

The objective is not to write for machines at the expense of people. It is to give both audiences information they can understand and verify.

3. Strengthen independent evidence of credibility.

Accurate professional profiles, relevant press coverage, public speaking, and other legitimate third-party references can help substantiate what a firm says about itself.

No agency can guarantee that an AI system will cite or recommend a firm, but a credible public record gives those systems better evidence to work from.

4. Measure what prospective clients are seeing.

Alongside rankings and referral traffic, firms should periodically test how major AI tools respond to realistic questions about their practice areas and locations.

They should record whether the firm appears, what sources are cited, whether the description is accurate, and whether inquiries ultimately become signed matters. AI-generated answers vary, so a single prompt is not a reliable performance measure.

Google's own guidance on AI features in Search reinforces the importance of foundational SEO, accessible text, and useful content. It does not require a special AI markup format to appear in AI Overviews or AI Mode.

A marketing priority, not a prediction

It would be premature to declare that AI assistants are now responsible for most legal-client acquisition. The research does not show that. Nor does the rise in bot requests prove that AI-driven recommendations convert into clients at a particular rate.

The more defensible conclusion is also the more important one: prospective clients are beginning their legal research inside AI systems, and those systems are becoming a new intermediary between law firms and the people they serve.

A law firm can no longer evaluate its visibility solely by how its website looks to a person or where it ranks on a traditional results page. It also needs to understand whether AI systems can discover its work and accurately represent its qualifications.

That is why marketing to machines belongs in a modern law firm's marketing strategy.

The ultimate audience remains human. But the path to reaching that person increasingly runs through AI.

Position Strategy Group works with founder-led law firms to assess and improve how they are represented in AI search. To request a complimentary AI visibility audit, contact Victoria at Position.

Research note: Traffic estimates cited here come from different providers, datasets, and time periods. “All bot traffic,” “AI traffic,” and “AI-referred visits” are different measurements. Results should not be combined into a single market-share estimate.

Citations

  1. “2026 Bad Bot Report.” Thales, 2026. cpl.thalesgroup.com
  2. “The State of Bot & Agent Security Report 2026: Key Takeaways.” DataDome, 22 Sept. 2026. datadome.co
  3. “AI Traffic Is Up 82%. Can You Tell Which Agents to Trust?.” DataDome, Sept. 2026. datadome.co
  4. “The 2025 Cloudflare Radar Year in Review.” Cloudflare, 15 Dec. 2025. blog.cloudflare.com
  5. “The 2026 State of AI Traffic & Cyberthreat Benchmark Report.” HUMAN Security, 26 Mar. 2026. humansecurity.com
  6. “Clio's 2026 Legal Trends Report.” Clio, 2026. clio.com
  7. “Google users are less likely to click on links when an AI summary appears in the results.” Pew Research Center, 22 July 2025. pewresearch.org
  8. “Overview of OpenAI Crawlers.” OpenAI, accessed 12 Nov. 2026. developers.openai.com
  9. “AI features and your website.” Google Search Central, accessed 12 Nov. 2026. developers.google.com

What does ChatGPT say when someone asks for you?

A positioning audit runs the questions your customers actually ask across the engines and shows who gets named today, and what it would take to change the answer.

Book a consultation  →