Legal information, not legal advice · Every entry verified to its primary source · Independent of any bar association

New Hampshire AI Ethics Rules for Lawyers

Last updated July 1, 2026 · First published July 1, 2026 · By MHSB Solutions (Research desk) · How this site is sourced

New Hampshire governs lawyers' AI use through three New Hampshire Bar Association Ethics Committee articles, each reviewed by the NHBA Board of Governors and published in 2024, tracked here as guidance because none is a formal ethics opinion. Ethics of Using Artificial Intelligence in Practice (May 15, 2024) holds that lawyers have no duty to adopt AI, but that Rule 1.1 competence requires understanding the tools, that outputs must be vetted before use on pain of Rules 3.3, 3.4, 4.1, and 8.4(c), that Rule 1.6 governs what client data reaches a tool, that Rules 5.1 and 5.3 make supervision of AI use a firm duty, and that Rule 2.1 keeps judgment human. Ethics of Drafting Documents with Artificial Intelligence (September 10, 2024) permits drafting with outside generative AI only with caveats, requiring Rule 1.4 consultation and Rule 1.6 informed consent before client information is disclosed to an outside tool, counseling against outside generative AI for pleadings, and flagging a possible future Rule 1.5 duty to use cost-reducing technology. Understanding Generative Artificial Intelligence and The Potential for Discrimination and Bias (October 30, 2024) reads Rule 1.1 to include understanding algorithmic bias and Rule 2.1 to require critical evaluation of output, while concluding that inadvertent AI bias alone does not violate Rule 8.4(g). New Hampshire had no formal NHBA opinion and no judiciary AI order at our verification, and an NHBA Special Committee on AI is developing further guidance.

Quick answer

  1. Authority: three NHBA Ethics Committee articles from 2024, each reviewed by the Board of Governors.
  2. May 15, 2024: no duty to adopt AI, but Rule 1.1 competence and output vetting if you do.
  3. September 10, 2024: client consultation and informed consent before outside tools see client data; avoid outside generative AI for pleadings.
  4. October 30, 2024: competence includes algorithmic bias; inadvertent bias alone is not a Rule 8.4(g) violation.
  5. No formal NHBA ethics opinion and no judiciary AI order at our checks.
  6. An NHBA Special Committee on AI, chaired by President-Elect Bob Lucic, is developing further guidance.

The official instruments

InstrumentTypeDateKey duties
Ethics of Using Artificial Intelligence in Practice Official guidance May 15, 2024 competence, verification, candor, confidentiality, supervision, independent judgment
Ethics of Drafting Documents with Artificial Intelligence Official guidance September 10, 2024 communication, consent, confidentiality, verification, supervision, fees
Understanding Generative Artificial Intelligence and The Potential for Discrimination and Bias Official guidance October 30, 2024 competence, bias monitoring, independent judgment

Guidance as a series, not a single document

New Hampshire answered generative AI the way a bar teaches rather than the way a court legislates: in installments. Across 2024 the New Hampshire Bar Association’s Ethics Committee published three practical-ethics articles, one on using AI in practice at all, one on drafting documents with it, and one on discrimination and bias, and each passed through the NHBA Board of Governors before publication, which separates the series from an individual author’s musings. The tracker lists all three as instruments, typed as guidance, because that is what they are: committee-written, board-reviewed positions without the formal-opinion label. New Hampshire had issued no formal NHBA ethics opinion on AI and no judiciary order when we ran our checks, so the series is the state’s attorney-facing authority, and the bar keeps a single hyperlinked index of its opinions and articles that makes the distinction easy to audit.

May 2024: adoption is optional, diligence about it is not

The opening article, Ethics of Using Artificial Intelligence in Practice, submitted to the Board of Governors on February 8, 2024 and published May 15, 2024, sets the frame. There is no ethical duty to adopt AI. Choosing to use it triggers Rule 1.1 technological competence, and the committee is specific about the failure mode: outputs must be vetted, with Mata v. Avianca as the standing example, because unvetted output can implicate Rules 3.3, 3.4, 4.1, and 8.4(c), the candor and honesty cluster the sanctions caselaw keeps enforcing. Rule 1.6 governs how tools handle client data, Rules 5.1 and 5.3 put AI use inside the firm’s supervision structure, and Rule 2.1 keeps professional judgment independent of the machine’s fluency. The article closes where every bar eventually lands, on firm-level safeguards and policies, the same architecture the policy template implements.

The second article, Ethics of Drafting Documents with Artificial Intelligence, reviewed by the board on June 7, 2024 and published September 10, 2024, draws New Hampshire’s sharpest line. Drafting with outside generative AI, tools that process data beyond the firm’s control, is permissible only with caveats: Rule 1.4 requires consulting the client, and Rule 1.6 requires informed consent, before client information is disclosed to an outside tool, and the committee counsels avoiding outside generative AI for pleadings altogether. Verification duties under Rules 1.1 and 3.3 cover everything drafted, supervision under Rules 5.1 through 5.3 extends expressly to nonhuman assistance, and generic drafting uses that expose no client information stay permissible. The article also plants a flag other states have mostly not: a possible future Rule 1.5 duty to use cost-reducing technology, fee ethics running toward AI rather than away from it, a thread the billing analysis follows nationally. The client disclosure guide covers the consent mechanics the committee assumes.

October 2024: bias, and the committee that comes next

The closing article, Understanding Generative Artificial Intelligence and The Potential for Discrimination and Bias, reviewed October 24, 2024 and published October 30, 2024, treats bias as a competence subject: Rule 1.1 includes understanding that models can carry discrimination into legal work. Its most practical holding is a boundary. Inadvertent AI-introduced bias is not by itself a Rule 8.4(g) violation, which requires purpose or animus, but Rule 2.1’s independent-judgment duty requires critically evaluating output rather than passing it through. The article commends ABA Formal Opinion 512, and our 512 breakdown shows how closely the national framework tracks New Hampshire’s series. What comes next is already chartered: an NHBA Special Committee on AI, chaired by President-Elect Bob Lucic, is developing further guidance, and whatever it produces will be verified against the bar’s own publications and logged in the changelog before this page changes. Until then the confidentiality breakdown and the supervision rules cover the two duties the series leans on hardest.

Frequently asked questions

Are the three articles binding on New Hampshire lawyers?

They are guidance, not formal opinions or rules, and this tracker labels them accordingly. But each was written by the NHBA Ethics Committee and reviewed by the Board of Governors before publication, which makes the series the bar's institutional position on AI, and the safest prediction of how a New Hampshire disciplinary authority would read the existing rules against an AI misstep.

Can I put client information into an outside AI tool in New Hampshire?

Only after the September 2024 article's two steps: consult the client under Rule 1.4 and obtain informed consent under Rule 1.6 before client information is disclosed to an outside tool. The committee also counsels against using outside generative AI for pleadings at all, and reminds lawyers that everything drafted must be verified before it is used.

Is biased AI output an ethics violation here?

The October 2024 article draws the line carefully. Inadvertent bias introduced by a tool is not by itself a Rule 8.4(g) violation, because that rule requires purpose or animus. But competence under Rule 1.1 includes understanding that models can carry bias, and Rule 2.1 requires critically evaluating output instead of adopting it, so passing biased output through unexamined is where trouble starts.

Is more formal New Hampshire guidance coming?

Machinery for it exists: an NHBA Special Committee on AI, chaired by President-Elect Bob Lucic, is developing further guidance. New Hampshire had no formal NHBA ethics opinion and no judiciary AI order when we ran our checks, and if either issues, this page and the changelog will say so once the primary source verifies.

Primary sources cited

Related guides

About the editor: MHSB Solutions, Research desk. MHSB Solutions is not a law firm. Everything on this site is legal information keyed to primary sources, not legal advice.

For licensed attorneys and firm operators. This site is legal information, not legal advice, and no attorney-client relationship is formed by using it. Rules change; verify against the primary sources linked on every page and consult a licensed attorney in your jurisdiction before acting.