How to Use AI for Legal Marketing Without Violating NJ RPC 7.1

Solo and small-firm attorneys in New Jersey are increasingly turning to AI to handle something they've always dreaded: marketing. Tools like ChatGPT, Claude, and purpose-built legal marketing platforms can now draft website bios, practice area pages, LinkedIn articles, and even client-facing newsletters in a matter of seconds.

The output is often surprisingly polished. And that polish is exactly the problem.

AI-generated legal marketing copy tends toward superlatives. It gravitates to phrases like "aggressive representation," "proven track record," "best results," and "we win." These aren't just hollow buzzwords — in New Jersey, they're potential ethics violations. And because the content sounds professional, attorneys sometimes publish it without the scrutiny they'd apply to a court filing.

That's a gap worth closing.

What NJ RPC 7.1 Actually Prohibits

New Jersey's Rule of Professional Conduct 7.1 governs attorney advertising and is broader than many practitioners realize. It prohibits communications that are false, deceptive, or misleading — including statements that create "unjustified expectations" about the results an attorney can achieve.

The rule doesn't require intent to deceive. If an AI drafts a homepage headline that says your firm "consistently delivers favorable outcomes for clients," that statement — published under your name, on your domain — is yours. The New Jersey Supreme Court's Committee on Attorney Advertising (CAA) has repeatedly flagged outcome-based language, comparative claims, and vague superlatives as problematic, regardless of how they found their way into print.

This matters because AI models are optimized to produce confident, persuasive prose. They don't know the difference between a compliant attorney bio and a marketing claim that runs afoul of RPC 7.1. That distinction is entirely your responsibility.

Three AI Marketing Outputs That Routinely Trigger Compliance Risk

1. Practice area pages with results-based language. Ask any general-purpose AI to write a personal injury practice page, and it will almost certainly mention client recoveries, favorable settlements, or winning records. Unless those results are fully substantiated and accompanied by the required disclaimer under NJ CAA guidance, they don't belong on your site. AI doesn't flag this. You have to.

2. LinkedIn posts framed as client success stories. AI is excellent at turning a vague prompt ("write a LinkedIn post about a good outcome I got for a client") into a narrative that implies your involvement was decisive. Even without naming the client, this kind of post can implicate RPC 1.6 confidentiality obligations and RPC 7.1 if the framing is misleading. The dual exposure is easy to miss when the copy looks clean.

3. AI-generated attorney bios that inflate credentials. Superlatives creep into bios almost by default. "Seasoned litigator." "Recognized authority." "Extensive experience." If the AI writes it and you don't verify it against your actual record, you've published a claim you may not be able to substantiate — which is precisely the kind of statement the CAA scrutinizes.

A Practical Review Workflow for AI-Generated Marketing Copy

You don't need to abandon AI marketing tools. You need a lightweight compliance gate before anything goes live. Here's what that looks like in practice:

Step 1 — Run a superlative sweep. Before reading for substance, do a word search for: best, top, proven, winning, consistently, guaranteed, leading, superior, extensive, recognized, aggressive. Flag every instance and ask: can I substantiate this with a verifiable fact? If not, rewrite it.

Step 2 — Strip outcome-based claims or add proper context. Any reference to past results must include NJ's required disclaimer (past results do not guarantee future outcomes) and should be specific enough to be verifiable. Vague references to "successful cases" or "millions recovered" without substantiation are exactly what the CAA targets.

Step 3 — Check the comparative language. AI frequently positions attorneys against competitors implicitly ("unlike big firms, we give you personal attention"). Under RPC 7.1, comparisons must be objectively verifiable. If it can't be measured, cut it.

Step 4 — Apply the "prospective client" test. Read the copy as someone who knows nothing about law would. Does any sentence create an expectation that you'll win their case or that your results will mirror past ones? If so, it needs a rewrite, regardless of how polished it sounds.

Step 5 — Date-stamp your review. Keep a simple log — a spreadsheet is fine — recording what AI tool generated the content, the date you reviewed it, and what changes you made. If the CAA ever inquires, this record demonstrates reasonable oversight.

The Bigger Picture

AI is a genuine force-multiplier for solo and small-firm marketing. A practitioner who previously couldn't afford a marketing agency can now produce consistent, professional content at low cost. That's a real competitive advantage — but only if the content clears the ethics bar.

New Jersey's RPC 7.1 was written long before generative AI existed, but its logic maps cleanly onto AI-generated copy: the medium of production is irrelevant. What's published under your name is your professional statement.

Build the review habit now, while the stakes are manageable. The attorney who publishes AI-generated marketing copy on autopilot is one CAA inquiry away from a very uncomfortable conversation — and that's the outcome no AI tool can draft its way out of.