ABA Formal Opinion 512 — Generative AI Tools

ABA Formal Opinion 512, issued in July 2024, is the foundational national ethics opinion on attorney use of generative AI. It does not create new duties — it maps existing Model Rules of Professional Conduct onto the specific risks generative AI introduces. Every COUNSEL pillar traces back to a duty articulated here.

Rules mapped by the opinion

• Rule 1.1 — Competence. Lawyers must understand the benefits and risks of the technology they use. AI competence is part of legal competence.
• Rule 1.6 — Confidentiality. Sharing client information with a third-party AI tool that retains, trains on, or exposes data implicates the duty. Informed consent is required absent contractual protection.
• Rule 1.4 — Communication. Material AI use that affects strategy, cost, or outcome may need to be communicated to the client.
• Rule 3.3 — Candor to the tribunal. Submitting AI-generated false authority is a candor violation. Verification is the cure.
• Rules 5.1 & 5.3 — Supervisory duties. Partners and supervising attorneys are responsible for the AI use of subordinate lawyers and nonlawyer staff.
• Rule 1.5 — Fees. Billing implications when AI compresses task time must comply with the reasonableness requirement.

The verification duty

The opinion's central operational holding: AI output is a draft, not authority. Attorneys must independently verify citations, holdings, statutory text, and factual claims against authoritative sources before relying on AI-generated content. The duty is non-delegable. Using a second AI to verify the first does not satisfy it.

The disclosure question

Opinion 512 declines to impose a uniform disclosure-to-client rule. Disclosure should track materiality — typo-fixing and grammar checks need not be disclosed; AI-assisted legal analysis that influenced strategy probably should be, particularly if the client would consider it material. Disclosure to the tribunal is governed by the rules of the specific court.

Confidentiality and informed consent

The opinion treats a public-facing AI tool that retains user inputs as a third party for Rule 1.6 purposes. Sharing confidential client information with such a tool requires informed client consent. Enterprise-grade tools with contractual data protection are evaluated under the existing "reasonable efforts" standard.

Fee implications

The opinion confronts the billing problem directly: if AI compresses a 10-hour task to one hour, the lawyer cannot ethically bill the 10 hours. The fee must be reasonable under Rule 1.5. Flat-fee and value-based billing models are not prohibited and may be the cleanest response.

How this anchors the COUNSEL Framework

The COUNSEL Framework was built to operationalize Formal Opinion 512 inside a working law practice. Each pillar maps to a duty articulated in the opinion:

• Competence ← Rule 1.1
• Oversight ← Rules 5.1 / 5.3
• Use Cases ← Rules 1.1 / 1.5 (matter-appropriate tool selection)
• Non-disclosure ← Rule 1.6
• Supervision ← Rule 5.3 (nonlawyer assistance)
• Ethics ← Rules 1.4 / 3.3
• Lifecycle ← Procurement / monitoring obligations implicit in 1.1

Practitioner takeaways

• Read Opinion 512 in full. It is short and operational.
• Adopt a written verification protocol and a written confidentiality protocol for AI use at your firm.
• Audit your engagement letters and fee structures for AI-era compatibility.

The full opinion text is available at the link above.