a series from Judy · five weekly letters ·

Letters on 512.

Five letters on ABA Formal Opinion 512 — the ethics opinion on generative AI in law. Tuesdays through May 19.

  1. Coming Tuesday · Apr 28

    · 13 min read

    Letter II · Confidentiality and the self-learning trap

    The bright line 512 quietly drew between two kinds of AI tool — what the ChatGPT Plus subscription on your laptop is doing with a client matter, why the engagement-letter paragraph fails the consent test, and the three-minute conversation that replaces it.

  2. Coming Tuesday · May 5

    · 12 min read

    Letter III · Fees, and the four-document file

    Why you cannot bill for the hour the AI saved, what the malpractice carrier will ask to see at renewal, and the four-document compliance stack — a one-page AI policy, a modified engagement letter, a client security explainer, a verification log — that answers both questions.

  3. Coming Tuesday · May 12

    · 12 min read

    Letter IV · Supervision, and the three conversations

    Rule 5.3 turns the tool into staff; the six-figure sanctions era has arrived to enforce it. Plus the three conversations — intake, annual review, post-near-miss — that do the supervising the documents only record.

  4. Coming Tuesday · May 19

    · 11 min read

    Letter V · Duties to the tribunal: the state overlays and where the ground is moving

    Where do you actually have to tell the court — and what is coming next? The last letter closes the circle: state overlays, standing orders, malpractice carriers, and a final word from me.

Every Tuesday, 9 AM Eastern, through May. A quiet correspondence in five parts. Published by Deci Ventures.

— Judy