What NJ Solo Attorneys Get Wrong About Who the Preliminary AI Guidelines Were Actually Written For

Most solo attorneys in New Jersey have read (or skimmed) the Supreme Court's "Preliminary Guidelines on New Jersey Lawyers' Use of Artificial Intelligence," issued January 24, 2024. A fair number treated it the way practitioners treat a new court administrative notice: scan for the operative language, file it mentally, move on.

That's a mistake, and it's a specific kind of mistake. The guidelines weren't drafted for a managing partner at a fifty-lawyer firm with a dedicated IT department and an in-house ethics counsel. The compliance apparatus assumed by large firms simply doesn't exist in a solo practice, which means the solo practitioner carries the same duties with none of the institutional scaffolding.

Here's where the disconnect matters most.

The guidelines apply the existing RPCs, not new ones

The Supreme Court was explicit: the preliminary guidelines don't create new rules. They map existing professional conduct obligations onto AI-assisted practice. That means NJ RPC 1.1 (competence), NJ RPC 1.4 (communication), NJ RPC 1.6 (confidentiality), and several others are already in play every time you open a generative AI tool on a client matter. The guidelines describe how those rules apply, not whether they apply.

For a solo attorney, this framing matters because it closes off a possible misreading. Some practitioners assumed that because the guidelines were labeled "preliminary," obligations were somehow in a holding pattern. They weren't. If you used an AI tool that transmitted client data to a third-party server without understanding the data handling terms, NJ RPC 1.6 was already implicated before the guidelines existed. The guidelines clarify the standard; they don't set a future effective date for it.

Verification isn't optional, and no one else is doing it for you

One of the clearest themes running through the January 2024 guidelines is that a lawyer must verify AI-generated output before using it. In a large firm, that verification step often gets layered: an associate runs the tool, a supervising partner reviews, a research librarian spot-checks citations. In a solo practice, you are all three people.

This isn't a small operational wrinkle. It means that any workflow you design around AI output has to build in a verification step you personally perform. A hypothetical solo family law attorney who uses a generative AI tool to draft a motion, reviews the output for style but not for citation accuracy, and files it has not satisfied the verification standard the guidelines describe, regardless of how plausible the citations looked on screen.

The guidelines don't specify exactly how verification must happen, which creates genuine judgment calls. What they do establish, by reference to the underlying RPCs, is that verification is a professional obligation tied to the lawyer's duty of competence. ACPE Opinion 701 (2006), which addressed electronic storage and cloud-based practice tools, established that reasonable care in selecting and monitoring technology services is part of that same competence standard. The 2024 AI guidelines extend that logic to generative tools.

The gap between "preliminary" and "permanent" is narrower than it looks

Practitioners sometimes treat the word "preliminary" in the guidelines' title as a signal that more lenient standards apply now, with stricter rules coming later. That's not an accurate reading. The Court described the guidelines as preliminary because AI technology is still evolving and permanent rules may follow once the technology and its risks are better understood. The underlying RPC obligations are not preliminary. NJ RPC 1.1, NJ RPC 1.6, and the supervision rules are in force now, with or without finalized AI-specific guidance.

What the preliminary status actually means for a solo practitioner is that you can't rely on the guidelines as a complete compliance checklist. They identify the general obligations but don't provide safe-harbor procedures. If you follow every suggestion in the guidelines and still transmit confidential client data to a tool with inadequate data protections, you haven't satisfied RPC 1.6 by virtue of guideline compliance.

Where solo practitioners should focus right now

The area where the guidelines create the most underappreciated risk for solos is the intersection of AI use and client communication. The guidelines note that transparency with clients about AI use may be appropriate or required depending on the circumstances. Because solos typically handle client relationships without a dedicated client intake team or a standard disclosure workflow, the decision about whether and how to disclose AI use in a particular matter falls entirely to the attorney, often in the middle of an already busy representation.

That's not a theoretical problem. A client who later learns their matter was handled with significant AI assistance, and was never told, has a reasonable basis to question whether the attorney met the communication standard under NJ RPC 1.4. The guidelines don't resolve every disclosure question, but they do put the attorney on notice that the question exists and is theirs to answer.

The most practical starting point for a solo attorney is a written, matter-by-matter habit: before using a generative AI tool on a client file, confirm you understand where the data goes, verify any output you plan to rely on, and decide whether disclosure is appropriate for that specific client and matter type. The guidelines won't do that work for you. They just make clear that you're the one responsible for it.

What the rule requires

The NJ Supreme Court's "Preliminary Guidelines on New Jersey Lawyers' Use of Artificial Intelligence" (January 24, 2024) apply existing RPC obligations to AI-assisted practice. They require attorneys to verify AI-generated output before use, to apply the competence standard of NJ RPC 1.1, to protect client information consistent with NJ RPC 1.6, and to consider whether client disclosure is required or appropriate under NJ RPC 1.4. ACPE Opinion 701 (2006) established that reasonable care in selecting and monitoring technology tools is part of an attorney's professional obligation. The guidelines do not create new rules, do not establish safe harbors, and do not defer existing RPC obligations pending finalized AI-specific rules.

What I recommend

Solo attorneys should stop reading the preliminary guidelines as a checklist and start reading them as a pointer back to the RPCs. For each AI tool you use on a client matter, identify which RPC obligations are implicated, verify you've satisfied them, and document that you did. Build verification into your workflow as a named step, not an assumption. For client disclosure, make a deliberate, documented decision on each matter type rather than applying a blanket policy or no policy at all. If you haven't reviewed the January 2024 guidelines alongside the text of NJ RPC 1.1, 1.4, and 1.6 at the same time, that side-by-side reading is the most useful hour you can spend on AI compliance right now.