Billing Rates, Candor, and Competence: What NJ's 2024 AI Guidelines Actually Say to a Solo Attorney Handling Their Own Cases

The NJ Supreme Court released its "Preliminary Guidelines on New Jersey Lawyers' Use of Artificial Intelligence" on January 24, 2024. The document runs fewer than 1,500 words. It has no footnotes, no lengthy commentary, and no defined enforcement mechanism built into its own text. And yet, for a solo attorney handling real estate closings in Hackensack or a two-person immigration firm in Newark, it sets the clearest picture New Jersey has of what practicing with AI is supposed to look like right now.

Most coverage of these guidelines has treated them as a supervision story, asking whether managing partners are watching their associates. That framing misses something. The guidelines speak directly to the individual attorney, regardless of firm size, and they do it by pointing back to the RPCs every NJ lawyer already carries.

What the guidelines actually say, paragraph by paragraph

The guidelines open by acknowledging that AI tools offer genuine efficiency gains but that they also produce errors, including fabricated citations and inaccurate legal analysis. From there, they lay out expectations organized around several existing professional duties.

On competence and verification. The guidelines state that lawyers who use AI must understand the technology well enough to evaluate its output. NJ RPC 1.1 requires competent representation, which the Supreme Court's guidelines tie directly to AI use: a lawyer cannot simply adopt AI-generated work product without independent review. The guidelines don't define "review" with a checklist, but the standard they invoke is practical competence, not theoretical familiarity.

On candor toward tribunals. The guidelines call out the risk of AI-generated citations that don't exist and make clear that submitting such material violates NJ RPC 3.3. There's no carve-out for "I didn't know the AI made it up." The guidelines put the verification obligation squarely on the attorney before anything goes to a court.

On confidentiality. The guidelines flag that inputting client information into AI tools raises confidentiality concerns under NJ RPC 1.6 and instruct lawyers to understand how any AI system handles, stores, or trains on data before using it with client matters. This isn't the ABA's framing; it's New Jersey's own, and it appeared in January 2024, months before ABA Formal Opinion 512 arrived in July of that year.

On fees. The guidelines note that AI-assisted efficiency may affect what a lawyer can reasonably charge, pointing back to NJ RPC 1.5. The exact billing implications are left to the attorney's judgment, but the signal is clear: if AI compresses a task from four hours to twenty minutes, billing four hours requires a justification that can survive scrutiny.

On supervision. Even for a solo attorney with no associates, the guidelines matter here, because solos often use virtual assistants, contract paralegals, or outsourced services. The guidelines connect AI use by non-lawyer staff to RPC 5.3, which places responsibility on the supervising attorney for that work product.

On disclosure. This is the part most NJ attorneys I've spoken with hypothetically as a thought experiment tend to overlook. The guidelines suggest that courts and, in some circumstances, clients may need to know when AI was used in the preparation of work. The guidelines don't mandate a specific disclosure form, but they signal that transparency is the expected default posture, not an optional courtesy.

What the ACPE has not said

One thing worth naming clearly: as of this writing, New Jersey's Advisory Committee on Professional Ethics has issued no opinion specifically addressing artificial intelligence. ACPE Opinion 701 (2006) covers electronic storage and cloud services generally, and it remains relevant background for thinking about data handling, but it predates generative AI entirely. Attorneys looking for a comprehensive NJ ethics opinion on AI tools are looking for something that doesn't yet exist. The Supreme Court's January 2024 guidelines are the controlling NJ-specific document, and they're explicitly labeled "preliminary."

That "preliminary" label matters. The Supreme Court signaled that more detailed guidance is coming. Attorneys who treat the guidelines as a final answer rather than a floor are building on an incomplete foundation.

The practical gap for a solo firm

Here's where it gets concrete. A solo attorney using an AI tool to draft a motion in limine faces at least four of the RPC touch points above simultaneously: competence (did I understand and verify what the AI produced?), candor (did I check every citation before filing?), confidentiality (did I put the client's name and case facts into a tool whose data practices I've never read?), and fees (does my time entry reflect actual attorney effort?).

The guidelines don't require an attorney to avoid AI. They require the attorney to practice with AI the same way they'd practice without it: carefully, honestly, and with full professional accountability for the output.

For a solo or two-person firm, that means building a short verification step into any AI-assisted drafting workflow, reading the data terms of whatever tool you're using, and keeping a contemporaneous record of what AI did and what you did. That last piece is mostly a self-protection measure, but it's a meaningful one if a client or court ever questions the work.

What the rule requires

The NJ Supreme Court's Preliminary Guidelines on New Jersey Lawyers' Use of Artificial Intelligence (January 24, 2024) require NJ attorneys to:

• Understand the capabilities and limitations of any AI tool used in their practice, consistent with NJ RPC 1.1.
• Verify all AI-generated content, including citations, before submitting it to a court, consistent with NJ RPC 3.3.
• Assess the confidentiality implications of inputting client data into AI systems, consistent with NJ RPC 1.6.
• Ensure that fees charged for AI-assisted work are reasonable, consistent with NJ RPC 1.5.
• Supervise any non-lawyer use of AI tools, consistent with NJ RPC 5.3.
• Consider whether disclosure of AI use is appropriate to the court or client.

ACPE Opinion 701 (2006) governs electronic storage of client files and cloud-based services under NJ RPC 1.6 but does not address generative AI specifically.

What I recommend

First, read the actual guidelines. The full text is short, and reading it yourself takes less time than reading most summaries of it, including this one.

Second, pick one AI tool you're currently using and find its data processing terms before your next client matter. Note in a file whether those terms are acceptable under your reading of RPC 1.6. If you can't find the terms, that's itself an answer.

Third, create a one-sentence entry in your matter notes any time AI assisted in drafting a document: what tool you used, what it produced, and what you changed or verified. This habit costs almost nothing and gives you a defensible record if anyone ever questions the provenance of your work.

Fourth, watch for the next round of NJ guidance. The "preliminary" designation in the January 2024 guidelines almost certainly means the Supreme Court or the ACPE is working on something more detailed. When it arrives, it will likely be binding in a way these guidelines arguably are not, and the attorneys who've already built clean habits will have far less adjusting to do.