Issue Archive

The Orthodoxy Journal

Reporting and analysis on AI in legal practice — sanctions, standing orders, malpractice exposure, and the evolving duty of competent supervision.

Latest · May 21, 2026

915 Opinions Later, "I Didn't Know" No Longer Works

A Central District of California court treats every attorney as already on notice of AI hallucination risk after citing 915 prior opinions on the same problem.

On April 28, 2026, a federal judge in the Central District of California sanctioned an attorney for a fabricated case citation submitted three times across two briefs — using LexisNexis+ and its Protégé drafting feature, not ChatGPT. The court cited 915 prior opinions on improper AI use and held that any attorney meeting the Rule 1.1 duty to stay current on technology is already on notice.

Read article →

Earlier issues

  • May 21, 2026

    When AI Writes Your Apology for Using AI

    On April 24, 2026, the Supreme Court of Alabama dismissed an appeal because the appellants' briefs contained too many fabricated citations to review. The attorney's reply-brief apology footnote — promising "the mistake will not recur" — cited two more cases that do not exist. Sanctions: $17,200, double appellate costs, co-signature requirement on future Alabama Supreme Court filings, and bar referral.

  • May 21, 2026

    A Broader Admonition to the Bar

    On May 5, 2026, the U.S. District Court for the District of Maine sanctioned plaintiff's counsel in Fuller v. Hyde School and paired the individual sanction with what it expressly called "a broader admonition to the Bar regarding the need for caution, verification, and transparency whenever generative AI is used in litigation." The order's distinctive contribution: non-disclosure of AI use as a Rule 3.3 candor problem analytically separate from the Rule 11 verification failure.

  • May 18, 2026

    The Four Levels of AI Citation Error

    An April 29, 2026 opinion from the Southern District of New York does what no prior AI sanctions order has done: it sorts citation errors into four categories and explains what verification each one requires. The most analytically useful document a court has produced on the subject of AI-assisted research.

  • May 18, 2026

    Bringing the Payne

    On May 5, 2026, the Supreme Court of Georgia suspended an assistant district attorney's privilege to practice for six months over AI-generated citations in three filings — and vacated a trial court's order that adopted the same fabricated authorities. Among the most severe AI-related practice sanctions yet imposed by a state supreme court.

  • May 9, 2026

    Why This Blog Exists — and What Orthodoxy Does About It

    Courts are not sanctioning attorneys for using AI. They are sanctioning them for not being able to prove they checked their work. That is the problem Orthodoxy was built to solve.