AI in NJ Law Firm Marketing: Where RPC 7.1 Draws the Line on Generated Ad Copy

Solo attorneys and small-firm practitioners across New Jersey have quietly started using AI to handle a task they've always loathed: marketing. A few prompts into ChatGPT or a tool like Jasper, and suddenly there's a month of LinkedIn posts, a refreshed website bio, and a catchy tagline for the firm's practice area page. It saves hours. It also creates risk that most practitioners haven't thought through.

New Jersey's RPC 7.1 prohibits attorneys from making false or misleading communications about their services. That rule doesn't care whether a human or an AI wrote the sentence. If a generated tagline overstates what the firm can deliver, the attorney whose name is on the website owns the ethics problem.

What AI Copy Gets Wrong About Legal Marketing

The core issue is that AI tools are trained to be persuasive. They default toward superlatives and confident claims because that's what performs well in commercial writing. That instinct is dangerous in a regulated profession.

Run a basic prompt like "write a marketing bio for a New Jersey criminal defense attorney" and most models will return something that includes phrases like "best possible outcome," "aggressive representation," or "proven results." Every one of those phrases can implicate RPC 7.1.

Under NJ RPC 7.1(a), a communication is misleading if it creates "an unjustified expectation about results the lawyer can achieve." The word "proven" is an almost automatic problem because it implies a track record the attorney can guarantee repeating. "Best possible outcome" is similarly shaky. "Aggressive" sits in a gray zone but has drawn scrutiny in other state bar guidance as an implied promise about strategy rather than a neutral description of style.

The New Jersey Supreme Court Committee on Attorney Advertising has consistently taken a conservative position on comparative and results-based language. Their guidance memos make clear that any claim that distinguishes the attorney as superior to others, without verifiable basis, runs afoul of the rule.

The Specific Failure Mode in AI-Generated Copy

There's a pattern worth knowing. When you ask an AI to write legal marketing content, it often produces what sounds like a client testimonial even when you didn't ask for one. Phrases like "clients consistently describe [Attorney Name] as..." or "past clients have found relief through..." appear frequently in generated output. Under NJ RPC 7.1(b), using client endorsements in attorney advertising is tightly regulated and requires specific disclosures. AI has no awareness of that obligation and will insert testimonial-style framing without flagging it.

Similarly, AI tools tend to fabricate or assume case results. If you prompt the model with context about your practice, it may generate language like "including complex multi-million dollar settlements" based on pattern-matching with similar attorney bios, not your actual record. That's straightforwardly false under RPC 7.1(a)(1) if it doesn't reflect your history.

A Practical Review Workflow That Actually Works

The fix isn't to stop using AI for marketing drafts. It's to treat every piece of AI-generated attorney copy as a first draft that requires a compliance pass before publication.

That pass should check for four categories of language: (1) results language, meaning any phrase that implies guaranteed or typical outcomes; (2) comparative superiority claims, including anything that positions the attorney as "top," "best," "leading," or "most experienced" without a verifiable, disclosed basis; (3) testimonial framing, even if no actual client is quoted; and (4) specialty designations, because NJ RPC 7.4 governs those separately and AI will freely describe an attorney as a "specialist" when they may not hold a certified designation.

If you're using AI for website copy specifically, it's worth cross-referencing the NJ Attorney Advertising Guidelines published by the Committee on Attorney Advertising, which were last revised in 2022. Those guidelines cover online content explicitly, and several of the scenarios described map directly onto AI-generated marketing errors.

One More Thing That Gets Overlooked

Metadata and page descriptions matter too. Many attorneys use AI to generate SEO meta-descriptions for their website pages, then never review them for compliance. A meta-description that says "New Jersey's top DUI defense firm" is a public-facing attorney communication. It's subject to RPC 7.1 the same way a Yellow Pages ad would have been twenty years ago.

The NJ Committee on Attorney Advertising accepts informal opinion requests from licensed New Jersey attorneys. If a specific piece of AI-generated copy falls in a gray area, that's the right channel to use before the content goes live, not after a grievance gets filed.