Category: Legal | Reading time: 5 min
The Classification
On 29 July 2024, the American Bar Association Standing Committee on Ethics and Professional Responsibility issued Formal Opinion 512, Generative Artificial Intelligence Tools. It was the first national ethics guidance on lawyers use of generative AI.
The opinion does not create new rules. It applies the existing Model Rules of Professional Conduct to a new technology.
Its most consequential holding is a classification: AI tools are nonlawyers for the purposes of Model Rule 5.3.
Why That Classification Carries Weight
Rule 5.3 governs a lawyer responsibilities regarding nonlawyer assistance. It requires that lawyers with managerial authority make reasonable efforts to ensure the firm has measures in place giving reasonable assurance that a nonlawyer conduct is compatible with the professional obligations of the lawyer.
Applied to an AI tool, that is a supervision obligation over a system that operates without a supervisor present for each action.
Opinion 512 directs managerial lawyers to establish clear policies governing generative AI use. It does not mandate a written document, but a written policy is the practical way to discharge the duty and to demonstrate it was discharged.
The Other Duties
Rule 1.1 competence requires a reasonable understanding of the AI tools in use, including their capabilities, limitations and how they generate output. That does not require technical expertise. It requires knowing that outputs can be fabricated, that training data has cutoffs, and that verification is not optional.
Rule 1.6 confidentiality requires evaluating whether inputting client information into a given tool is consistent with confidentiality obligations before doing so.
The opinion also addresses communication with clients and reasonable fees consistent with time actually spent when AI is used.
On verification, the opinion declines to set a single standard. The required review is factually specific and depends on the tool and the task. Document review warrants more independent review than idea generation. In every case, AI cannot substitute for competent legal work.
The State Layer
More than thirty five state bar associations have issued AI guidance as of March 2026. Over forty have issued ethics opinions.
The New York City Bar issued Formal Opinion 2024-5. Florida issued Ethics Opinion 24-1 in January 2024, among the first. California published proposed rule amendments in March 2026.
The convergence across these opinions is consistent: consumer AI products are presumptively insufficient for matters involving confidential client information without additional safeguards, and attorneys remain personally responsible for AI assisted work product.
Courts Are Separate
The New York State Unified Court System adopted 22 NYCRR Part 161, a system wide AI policy for all Unified Court System courts, effective 1 June 2026. It establishes disclosure and certification requirements for AI generated filings.
Court obligations run parallel to bar obligations and are enforced differently. A filing that satisfies Rule 1.1 competence can still violate a court disclosure rule.
The Gap
Adoption among legal professionals moved from roughly 11 percent in 2023 to 30 percent in 2024 by the ABA own survey, with other measures placing it far higher.
Policy adoption did not keep pace. The exposure sits in the gap between using generative AI and documenting the supervision the Model Rules require over it.
Rule 5.3 asks what measures were in place. Answering that after an incident is considerably harder than recording it as the work happens.