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.

Lemon Law lawyer Michelle Fonseca-Kamana discusses the seismic shifts in California lemon law—from the Supreme Court's decision in Rodriguez v. FCA US LLC (October 31, 2024) 17 Cal.5th 189 that effectively eliminated most used car claims, to the explosion in case filings (from 4,500 in 2015 to over 22,000 in 2023), to new legislative reforms under AB 1755 and SB 26 that impose strict timelines and mandatory pre-suit notice requirements.

Michelle also shares how she pivoted from in-person networking to social media marketing during the pandemic, built a practice around one-way fee-shifting statutes, and navigates the asymmetric litigation battlefield against billion-dollar manufacturers.

Highlights:

• Rodriguez v. FCA's impact on used-car protections: The Court limited manufacturer liability to certified pre-owned vehicles, leaving used-car buyers without recourse even when cars remain under manufacturer warranty.
• Why lemon law filings quintupled: Despite expectations that Rodriguez would reduce litigation, filings increased fivefold (2015-2023) due to declining vehicle quality, PI firm diversification, and political headwinds.
• New procedural requirements under AB 1755 and SB 26: Effective 2025, consumers must send pre-suit demand letters, wait 30 days, retain the vehicle, meet hard deadlines (one year after warranty expiration or six years from delivery), and navigate an "opt-in" system.
• One-way fee-shifting as equalizer: Song-Beverly allows consumers to bring claims without paying fees—manufacturers pay all costs if consumers prevail.
• Social media as practice-builder: Michelle built her practice through bilingual video content on Instagram, TikTok, and YouTube, generating clients and referrals without traditional marketing.
• Documentation mistakes: The biggest error is failing to keep itemized repair orders and contemporaneous complaints—gaps that become fatal under new requirements.

Tune in for insights on asymmetric consumer litigation, the intersection of statutory interpretation and real-world consequences, and how procedural reforms quietly reshape substantive rights.

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.

You have to literally disobey an order in California to be held in contempt. But federal courts are a little more touchy-feely: they will find a contempt for violating the “spirit” of their orders. Tim and Jeff compare the Ninth Circuit's contempt finding against Apple in the Epic Games dispute, and a state litigant who got around a visitation-time order but without violating the letter of the order, so no contempt.

Meanwhile, a CEQA plaintiff that won at the Court of Appeal—only to be reversed by emergency legislation and the Supreme Court—learned the hard way that "prevailing" on the law as written means nothing if the Legislature rewrites the rules mid-case.

Key points:

• Contempt requires literal violation in California, not just bad faith. But in federal court, violating the “spirit” of an order is contempt.
• Legislative abrogation torpedoed $1.2M in CEQA fees: Plaintiffs in Make UC a Good Neighbor v. Regents won significant CEQA victories establishing that crowd noise and alternative locations must be analyzed—then watched the Legislature pass emergency legislation abrogating both holdings. After the Supreme Court reversed, the Court of Appeal denied nearly $1.2 million in private attorney general fees, calling the prior opinion "smoldering ruins, not citable precedent." The court held plaintiffs weren't "successful parties" because they failed to halt the project, even though they vindicated principles under the law as it existed when filed.
• Ninth Circuit discovery ruling survives en banc review: The court declined to rehear the Trump administration's challenge to a discovery order requiring production of federal reorganization and layoff plans, rejecting executive privilege claims without requiring plaintiffs to show bad faith. Judge Bumatay's dissent warned of a "binding dicta trap" where the panel's comments on what qualifies as deliberative could become binding precedent.
• California Supreme Court limits Public Records Act obligations: Superior Courts can issue declaratory relief even after documents are produced if the dispute is likely to recur, but the Public Records Act does not impose a statutory duty to preserve documents a public agency identifies as exempt.

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.