Artificial intelligence is now a routine part of legal practice, and business lawyers must follow the same ethical rules that govern traditional work when they turn to generative AI tools.
Ethical duties remain unchanged
The ABA Formal Opinion 512 makes clear that AI does not create a new exemption from existing professional responsibilities. Lawyers must still provide competent representation, protect client confidences, communicate clearly, supervise staff, pursue only meritorious claims, maintain candor with courts, and charge reasonable fees. The opinion stresses that “to ensure clients are protected, lawyers using generative [AI] tools must fully consider their applicable ethical obligations.”
Model Rule 1.1 requires competence, but the rule does not demand that attorneys understand the technical code behind AI systems. Instead, a working knowledge of how the tools function is enough to show that lawyers are keeping pace with technology that affects client service. The panel at the ABA Business Law Section’s Spring Meeting in Atlanta urged lawyers to supplement AI training with input from information‑security and privacy specialists, expanding their perspective beyond pure technical expertise.
Confidentiality and data handling
When lawyers feed data into generative AI, Model Rule 1.6 obligates them to preserve confidentiality. This means mapping data flows, confirming that third‑party vendors do not reuse client inputs, and vetting vendor policies for privacy risks.
Courts are still sorting out how privilege and work‑product doctrines apply to AI‑generated material. Cases such as Warner v. Gilbarco, Inc. and United States v. Heppner illustrate that judges may treat AI output as ordinary work product, but they also warn that inadvertent reliance on inaccurate citations can create cascading problems. Because AI “amplifies both good and bad lawyering,” lawyers must verify any AI‑produced content before filing it with a tribunal.
Related: Court Rules in Favor of OppFi
Billing practices also evolve with AI use. Model Rule 1.5 requires fees to be reasonable, and while granular hour‑by‑hour detail is not mandatory.
In practice, these requirements mean that a business lawyer who relies on AI for contract drafting must still review each clause for accuracy, ensure that the AI platform does not retain confidential terms, and inform the client that the draft was AI‑assisted. Failure to do so could breach both confidentiality and competence standards.
For many firms, the shift involves building internal protocols that track AI inputs, retain audit logs, and define who is authorized to supervise AI‑generated work. They mirror the oversight duties already expected of paralegals and junior associates, simply extending the chain of responsibility to include the software itself.
Compliance is essential.
