California is the largest common-law jurisdiction where appellate courts don't follow each other—and it happened by accident. In Part 1 of this two-part episode, Michael Shipley explains how Bernard Witkin’s treatise reflections on case dicta became binding law, why the federal circuit model works differently, and what the rule costs practitioners and trial judges every day.

Key points:

• The Witkin origin story: No California Supreme Court decision actually establishes the no-horizontal-stare-decisis rule. It developed through dicta, then appeared in Witkin's first edition—which courts then cited as authority.
• The federal contrast matters for forum strategy: In the Ninth Circuit, Miller v. Gammy binds all panels within the circuit to follow the first published decision on an issue. California trial courts, by contrast, face conflicting appellate authority and must guess which rule the Supreme Court would adopt under Auto Equity—a burden one trial judge called being "appointed to the Supreme Court for temporary purposes."
• Stare decisis isn't jurisdictional (probably).
• Unpublished opinions create tension.
• The pros: California's rule allows multiple perspectives on emerging issues and prevents the first Court of Appeal decision from locking in statewide law before the Supreme Court weighs in.
• The cons: The rule creates uncertainty, burdens trial courts, and leads to inadvertent inconsistencies on procedural issues too minor for Supreme Court attention—splits that can persist for years or even decades. (In anti-SLAPP law, it took 13 years before Baral v. Schnitt decided how to handle mixed causes of action.)
• Publication practices hide the problem: Many conflicts never surface because courts strategically leave decisions unpublished, masking the frequency of divergent reasoning and making the appellate landscape harder to navigate.

Listen to Part 1 now for the full discussion on how California got here and what it costs practitioners—then tune in to Part 2, where Shipley covers forum shopping, the anti-SLAPP mixed-causes-of-action case study, and his proposed reform: precedential transfer.

In the first half of their conversation with James Mixon, Managing Attorney at California's Second District Court of Appeal, Tim Kowal and Jeff Lewis ask what is healthy AI use, and unhealthy use? To help organize—yes! To replace judgment—no! Tip: When an attorney does not read AI output before filing a brief, expect sanctions.

James draws on his role on the judicial branch AI Task Force and his monthly Daily Journal AI column to provide a practical roadmap for responsible AI use—from crafting effective prompts to avoiding the automation bias that has led to attorney sanctions across the country.

Key points:

• Treat AI as an on-demand legal treatise, not a research tool: Mixon explains how AI excels at providing background information and organizing legal concepts into digestible narratives—making it ideal for learning complex areas quickly—but should never replace verified legal research or case citation.
• The "Daedalus Doctrine" framework offers a middle path: Drawing from Greek mythology, Mixon warns against flying too high (reckless AI adoption) or too low (ignoring AI entirely), urging lawyers to use AI thoughtfully while maintaining personal judgment and verification responsibilities.
• Effective prompting is critical: Never use open-ended commands like "enhance this brief"—instead, tell AI exactly what you want and ask it to flag changes in italics or bold so you can review selectively.
• Hallucinations remain the biggest risk: Recent sanctions cases show attorneys asking ChatGPT to verify its own fabricated cases—a fatal error that demonstrates why every citation must be independently confirmed.
• Courts aren't using AI for decision-making: Current California court policy prohibits AI use "in any way that would touch a decision" to preserve public confidence over efficiency gains.
• AI works best for background learning: Mixon describes using AI to create narratives and explanations that make legal concepts stick—transforming dry doctrine into memorable stories, like having a personalized treatise writer available on demand.

Tune in to learn how to harness AI's power for legal background and organization without falling into the traps that have cost other attorneys their credibility—and thousands in sanctions.

Is your AI training data biased? And is using AI-generated reasoning plagiarism?

James Mixon, Managing Attorney at California's Second District Court of Appeal, covers troubling topics on how lawyers should, and should not, use AI. In this second part of Tim and Jeff’s conversation, James discusses how we can detect and counteract bias baked into training data. And what happens when trial judges unknowingly sign orders containing fabricated cases?

Key points:

• Legal reasoning isn't “creative” work—it's problem-solving: When we use words to solve problems, it should not be considered “plagiarism.”
• Bias detection requires active testing: AI models trained on historical data replicate past discrimination, particularly in employment, housing, and finance cases. James suggests an interesting experiment to try in your next research prompt.
• Alternative dispute resolution raises new questions: California bill Umberg 643 bars using AI for arbitration decision-making, reflecting concern that people signing arbitration agreements assume human decision-makers. If contracts explicitly state "AI dispute resolution," that might be acceptable—but not if buried in fine print.
• When should you disclose your AI use? Depends on where the use falls on a spectrum of “organization” and “discretion/judgment.”
• Trial court orders present a growing risk: Judges should strip proposed orders down to essentials: parties, motion, ruling, hearing date.
• AI lacks "ethos"—for now: AI currently can't replicate the credibility and reputation that make people trust human experts. This may change as AI systems develop track records, but for now, judicial decision-making requires the human judgment that builds public confidence in courts.
• Looking backward creates civil rights risks: AI trained on historical data is inherently conservative. Some models predicted Brown v. Board of Education would be affirmed based on precedent—a stark reminder that purely probabilistic decision-making can't account for moral progress.

What AI uses do you find most attractive—and the most troubling?

Disclaimer: The views expressed by our guest, James Mixon, are his own and do not reflect the official position of the California Court of Appeal or the California Judicial Branch. AI technology and legal standards are rapidly evolving, listeners should verify current rules and consult qualified attorneys before implementing AI tools in their practice. Attorneys must independently verify all legal citations and comply with applicable rules of professional conduct.