People v. Brim (Cal. Ct. App. - June 17, 2026)

It has been nearly a week since the Court of Appeal has issued a published opinion, so I was glad to read this one by Justice Richman today. It's an extremely good one. Well written, well argued, and persuasive on the merits. Really well done.

My only marginal critique is that I felt as though the opinion could have given the deceased victim here a touch more respect. The victim -- Oliver Williams -- was stabbed to death by the defendant (Jermaine Brim) during a fight on a BART train that was instigated when the barefoot defendant attempted to get some shoes from someone sleeping on the train. Mr. Williams stepped in to confront Mr. Brin and ordered him to leave the train; Mr. Brin did so, but then returned, a scuffle ensued, and ultimately Mr. Brin stabbed Mr. Williams to death.

Maybe the case involved a second degree murder, or maybe it involved voluntary manslaughter. That is what the published opinion is all about, and rightfully so.

But I would have preferred that the opinion, at least once, describe the deceased victim, Mr. Williams, as a good Samaritan. Because that's surely what he was. For better or worse, he was trying to protect someone else on a BART train. He didn't have to do that. Many -- perhaps most -- other people in that situation would have simply turned away, or watched the events transpire helplessly.

Mr. Williams didn't do that. He tried to help. And he ultimately paid for that decision with his life.

That deserves, in my mind, at least a little respect. Were it me, I would have expressly given it.

Quinteros v. Harbor Dist. Co. (Cal. Ct. App. - June 11, 2026)

No one likes to get sanctioned. No one likes to have their name sullied in public, or be compelled to pay $5,000 to the other side and another $1,000 to the court.

But sometimes, you just have to let it be. Rather than appeal the thing and result in a published opinion like this one.

It's yet another opinion about AI-generated briefs. (Anyone starting to get the message on that?) This time from a firm that used a contract attorney and didn't check his work -- amongst other problems.

Not only does the law firm and its attorneys lose the appeal, but it results in an opinion that includes the following representative content:

"The [trial] court compared quotations from the opposition to existing legal authority and stated, LLG “repeatedly and seriously misrepresent[s] the holdings of these and other cases” and “blatant misrepresentations appear throughout the brief.” The court pointed out, “the brief contains no fewer than eight fabricated quotations that purportedly appear in cases cited by” LLG and stated, again in bold font, “Literally every other purported quotation from a case in the brief is similarly fictitious.” The court expounded: “If these fabricated quotations were created by the use of a generative artificial intelligence tool such as ChatGPT, they are even more insidious than quotations from ‘hallucinated’ cases because they are attributed, falsely, to actual reported (if miscited) cases, which renders them more difficult and burdensome to detect." . . .

At the July 11 hearing on the OSC, Schelly and Badawi appeared in person as the court had directed. The court began the hearing by explaining the reason for the OSC: “The brief contained inaccurate citations, misrepresentations of California law, and remarkably eight different purported quotations from cases that are as far as the court can determine, entirely fabricated, invented, they don’t exist. [¶] They don’t appear in any of the cited cases. They don’t appear in any reported California case authority.”

Speaking on LLG’s behalf, Schelly apologized to the court and stated, “we’re all absolutely embarrassed, dismayed, upset that something like this has occurred involving our firm.” The court responded, “unfortunately, Mr. Schelly, the damage is done, but in the sense that both the court and your opponent had been put [through] the burden of reading and responding to a brief that turned out to be dishonest and—literally, almost literally, at every page. [¶] This is the worst example of misconduct by a lawyer that I think I’ve ever seen since I’ve been on the bench.”"

Yeah. Might have wanted to just take your lumps and go home on that one.

People v. Russo (Cal. Ct. App. - June 12, 2026)

The Court of Appeal decided to publish this (originally unpublished) opinion, which means that the following opening paragraph now gets to live on in perpetuity:

"[I]n February 2024, defendant’s mother told defendant to leave her property. Defendant responded by punching his dog approximately 20 times, calling it racial slurs, and dragging it down the street."

Dude: Why take it out on the dog?! (Also: Calling your dog racial slurs? Really? Does that even make any sense?)

In re Marriage of Capos (Cal. Ct. App. - June 12, 2026)

The Court of Appeal amends its opinion, inter alia, to add the following footnote (emphasis in original):

"The trial court’s math appears to be wrong, and the total amount awarded ($768,748.65) does not match the total amount LaShelle requested ($768,478.65), but neither party mentions these discrepancies, so we do not either."

To be honest, I do that whole "switch the numbers" thing on occasion as well.

County of Del Norte v. Britt (Cal. Ct. App. - June 11, 2026)

A guy who owns a house dies intestate. No one else is on the deed, and no one claims the house. Vagrants (predictably) take it over, and it gets ugly. The County then declares the house a nuisance, gets a receiver appointed, evicts the squatters, and hauls off fourteen tons of trash, junk and debris.

All this makes sense. Good government at work.

With the following caveat.

The dead guy has two sisters. They have zero interest in the house. They never opened up a probate estate and never sought to get any of their brother's assets. The County asks them if they care at all about what happens to the house. The sisters say "No." It was their brothers'. He's dead. Deal with it however you'd like. Burn it down, for all they care.

After evicting everyone and cleaning out the trash, the receiver then rehabilitates the house, finds a buyer, sells it, and uses the proceeds to off the mortgage as well as the costs incurred by the County. But those funds weren't sufficient to fully pay the receiver's own fees.

So the receiver then asks the court to make the sisters pay his fees.

The sisters hire a lawyer, who essentially says "Hey, we didn't do anything here. It's not our house, we never wanted it, we never filed anything, and we shouldn't have to pay fees for something of which we had utterly no part." The sisters are also miffed because the receiver could have gotten a super-priority lien to make sure the receiver's own fees got paid first, but inexplicably failed to do so.

The trial court agreed, and didn't hold the sisters responsible. But when the sisters then moved to recover their fees, the trial court refused, holding that they were not recoverable.

The Court of Appeal reverses, holding that they were.

There's no way anyone should have gone after the sisters in the first place. They weren't responsible. It wasn't their house, they didn't want it, and they shouldn't be liable for something just because they're the closest relatives to someone who dies intestate. Just like someone who dies with a credit card debt, that's exclusively the dead person's problem, not the relatives'. (At least if the relatives, as here, don't want any benefit from the estate.)

I'm happy the sisters got their fees. Both in the trial court and on appeal.

 

In re Tung Trust (Cal. Ct. App. - June 9, 2026)

There hasn't been a ton of interesting action in the Ninth Circuit or California Court of Appeal lately, but I've been thinking a little about this opinion ever since it came out the day before yesterday.

It's not that I disagree with either the result or the reasoning. It's actually an outstanding opinion by Justice Fueur, and I'm totally down with everything she says. Indeed, in my mind, the case is a great example why we have appeal as a matter of right. Because sometimes, a single judge gets it wrong, and it makes sense to have three smart judges take at least a quick look at the thing (and sometimes a deeper look) to correct any errors that might have transpired. Even if in the majority of cases, the result ends up getting affirmed.

Here, by contrast, there's a reversal. And rightly so.

To me, the ruling in the trial court is a great example of how, on occasion, judges get too bogged down in doctrine and purported efforts to confine themselves to contractual (or statutory) language and, in doing so, leave common sense behind.

The question here is how to distribute the assets of a revocable trust once the settlor dies. In 2011, an unmarried mother, Ya-Ching Tung, established a trust that designated as successor beneficiaries her three adult children. Typical. Moreover, the trust contained all the usual boilerplate paragraphs about additional property that might be added to the trust, the 30-day survivor provision, a no contest clause, etc.

In 2016, the oldest of the three children died, leaving behind his three kids. That son was supposed to get the majority of the property in the trust. In 2019, the settlor (the mother) died, and shortly thereafter, in 2020, one of the two daughters in the trust also died, without issue. That left alive only one of the three original children -- the one who was originally supposed to get a tiny fraction of what was in the trust (less than 1/6th) -- alongside the three grandkids, the children of the eldest son.

The surviving daughter then becomes the trustee of the trust, and petitions to have . . . get this . . . all of the trust property given to her, because she was the only direct child still alive. She admits that California law generally says (of course it does!) that surviving children generally stand in the shoes of their parents if the parents are beneficiaries and die. But the surviving daughter says that doesn't apply because the boilerplate 30-day survivor provision of the trust "demonstrates a contrary intent" and means that the three grandchildren are totally cut out, leaving the daughter as the sole beneficiary.

The trial court agrees with her. The Court of Appeal reverses.

As I said, I'm totally on board for Justice Feuer's reasoning and result. [My bad on the original spelling, though -- sorry about that!] Spot on right.

But I was also fundamentally displeased with how the trial court went about its task here. Yes, we're interpreting a written document -- a trust. Yes, we generally follow what it says, and look to what the terms mean. Yes, I understand why a judge might (wrongly) view the 30-day survivor provision as perhaps saying that if one of kids predeceases her, that kid is treated as dead. (Though, truthfully, that provision is really just intended to avoid the problem of figuring out "who died first" in a situation in which the settlor and one of the beneficiaries gets offed in the same incident.)

But here's the thing. There's zero reason to think that that's how the settlor ACTUALLY intended her trust to be distributed. There's ZERO reason why the grandmother would purportedly want the child who she gave the least amount under the trust (less than a sixth) to have the whole thing, while leaving her grandkids out in the cold merely because their mother had the misfortune of dying first. That's not how people actually act. I'm a thousand percent confident that was not, in fact, the settlor's intent. Yet that's what the trial court ended up deciding.

Common sense should matter. Actuating the intent of the written instrument should matter. That's got to play a role. Even if you think -- and especially if you think mistakenly -- that the words in the document technically or literally mean something else.

Text matters. So does intent.

And yeah, yeah, yeah, I know the whole parole evidence rule and the like. I understand that principle. Ditto for statutory interpretation, which follows the same general principle.

But written text is often more ambiguous than what we facially perceive it to be. Particularly when, as here, we've got a very reasonable, common sense interpretation of what that written text was intended to do.

Short version: Don't lose the forest for the trees. Don't forget what we're actually trying to do in cases like this one. Yes, we're reading text. But we're also very much trying to effectuate intent.

Don't leave common sense behind.

Nelson Misconduct Complaint (9th Cir. - June 8, 2026)

There we no published opinions yesterday, but on a different website, the Ninth Circuit issued this Order, which identified a judicial misconduct complaint about Judge Ryan Nelson.

The complaint stems from a parking lot dispute in Idaho Falls in April involving Judge Nelson. Judge Nelson was ultimately charged with misdemeanor battery and malicious injury to property as a result of that incident.

I was ultimately able to find the surveillance video here. Feel free to watch it yourself, but it very much looks to me like one party to that incident was absolutely the aggressor, and one can well see why charges might well have been brought against Judge Nelson here.

It's most definitely not a good look for a federal appellate judge.

Pizzuto v. Valley (9th Cir. - June 5, 2026)

It doesn't particularly bother me when death penalty opinions recite the facts of the underlying crimes in death penalty cases even though those facts are totally irrelevant to the issue at hand. Particularly when (as here) the federal panel elects to affirm the death sentence, I understand the persuasive effect of including those facts. Plus, as a general matter, it keeps the victims in mind. Opinions that reverse death sentences typically do the same thing with respect to the childhood traumas often endured by the defendant, so there's also a rough equality there. To a degree, everyone's playing in part on everyone's heartstrings.

But, for me, intellectual honesty requires that one at least recognize the other side of the equation. Unless the opinion only talks about the relevant facts, once you've decided to add technically irrelevant pro-prosecution (or pro-defendant) facts, one should at least fairly present some details about the other side.

I mention that in connection with today's Idaho death penalty opinion by Judge Bennett. Gerald Pizzuto claims that the Governor of Idaho violated his constitutional rights by not accepting the recommendation of the Idaho Commission of Pardons and Parole to commute his death sentence to life in prison. The Ninth Circuit's opinion affirms the decision of the district court to dismiss Mr. Pizzuto's habeas petition in that regard.

As is usual in these types of cases, Judge Bennett's opinion recounts the underlying (heinous) facts of the cold-blooded murders at issue even though they're technically irrelevant to the constitutional claims raised in the petition. The opinion also highlights how consistently Mr. Pizzuto has (understandably) fought to overturn his death sentence, and how long he's been on death row at this point: 40 years.

Again, I have no objection to that.

But the opinion is simultaneously bereft of any of the facts on the other side. One searches the opinion in vain, for example, to understand why the Idaho Commission of Pardons and Parole -- not exactly a hotbed of liberalism -- would even possibly recommend that Mr. Pizzuto's sentenced be commuted to life without the possibility of parole. One is left to entirely speculate on that point; there must be at least some facts on that score, but they're deliberately omitted. And given that at issue in the opinion is at least tangentially related to whether it was rational for the Governor to discard the Commission's recommendation, leaving those facts out is particularly noticeable.

Some of those facts are that Mr. Pizzuto (1) is now 70 years old, (2) has severe Type 2 diabetes, with damage to his arms and legs that has confined him to a wheelchair for years, and (3) is in hospice with advanced, terminal bladder cancer (alongside heart disease and COPD).

Which helps at least little to understand why the Commission recommended that the state leave this suffering old man to his fate in prison rather than continue to push forward with his execution.

One need not come to a conclusion regarding whether the Commission was right or wrong in this recommendation. But a fair presentation of the facts might elect to contain at least contain a sentence or two describing why the underlying facts at issue are what they are.

C.F. v. Alternate Family Services (Cal. Ct. App. - June 4, 2026)

I mean, I get it. Appellate courts don't typically reconsider a jury's allocation of fault. It's a fact issue. Which is why the Court of Appeal leaves it alone here, and affirms a $24.7 million award.

But allocating 60% of the responsibility for the molestation to the foster agency who screened the parent, and only 35% to the father who actually did the molestation? (Plus 5% to the mother.)

Come on. No one outside the legal system would actually find such an allocation rational.

Phillips v. Volvo Penta (Cal. Ct. App. - June 4, 2026)

The Court of Appeal holds today that when a boat engine manufacturer warranties the boat engine and then replaces the engine when that engine no longer works, that's sufficient under the Song–Beverly Consumer Warranty Act -- the engine manufacturer doesn't have to replace the entire boat.

To which my reaction was, and remains, this:

Duh.

Lnu v. Blanche (9th Cir. - June 3, 2026)

I normally don't discuss many of the "AI hallucination" opinions that regularly come down these days. For one thing, they're fairly straightforward (and uniform), and all basically say "It's okay to use AI, but if you do, you must make sure to thoroughly check the stuff to make sure it's right, and if you don't, you can get into a ton of trouble." Common sense, and entirely right, so I don't have much to add to it. Plus, generally, these opinions get a lot of mainstream press and attention anyway, so it's not like it's really important for me to highlight them and express my agreement with what they say.

Today's opinion from the Ninth Circuit, by contrast, is different. It's definitely worth mentioning.

First off, it's one of the very few Ninth Circuit opinions on the subject, so it's important for anyone who's litigating in that forum. Second of all, it contains a very strong claim about what an attorney's duties are with respect to submissions that they sign -- one that might, in fact, be too strong. So it's very much worth talking about.

The first paragraph of the opinion contains a good summary of what went down in this particular case:

"Attorneys Mike Singh Sethi and William Rounds filed briefs in this Court with multiple nonexistent cases, misattributed quotations, and gross misrepresentations of real cases. Sethi and Rounds claimed that the errors were the product of innocent typographical mistakes. And they repeatedly denied the possibility that generative artificial intelligence (“AI”) might have produced the errors. Having identified other cases in which Sethi or Rounds filed briefs that presented similar problems, we ordered them to show cause why they should not be sanctioned, suspended, or disbarred from practice before this Court. We have considered their response, and we now impose discipline."

Cutting to the chase, the attorneys here ultimately get sanctioned ($2500 each); suspended from the Ninth Circuit for six months; reported to the Bar; and ordered to provide a copy of the opinion to each of their clients, opposing counsel, and the presiding judge in every tribunal in which they have a case. Ouch.

I'm not sure that anyone is going to decry what happened to these particular lawyers (other than their friends and family, of course). Among other things, the opinion stands as yet another reminder that when you're caught for something like this, don't try to defend or deflect. Just go full mea culpa. You can tell that the panel here was very much not happy about the way the attorneys here responded to the court's investigation into these events. (Lines like this one permeate the opinion: "Instead of transparency, and despite our Order to Show Cause, Sethi has apparently chosen to engage in more subtle subterfuge.")

What I really wanted to focus on, though, is the Ninth Circuit's overall articulation of what it means to be a lawyer who signs a brief in that tribunal -- whether generated in whole or part by AI, or otherwise.

We obviously all know about Rule 11 and what it means for us to sign a brief. We're responsible for what's in it. That much is clear, and undisputed.

When we write the briefs ourselves, there's no real complexity. We're in charge, and we have to make sure that everything we say is entirely accurate.

But in the real world, lawyers often delegate. Associates write first drafts. Subordinate partners take their turn. Someone high up gives it an additional read. Someone ultimately signs the thing; generally, someone fairly high up the food chain.

All of that is fine as well.

What struck me about today's opinion, though, is how the Ninth Circuit articulates the duties of anyone who signs the brief. Citing the opinion from the California Court of Appeal, here's what the panel says:

"But a competent and diligent attorney must do more than prompt generative AI, check that the citations provided by the AI are real and the subject matter roughly on point, and call it a day. Cal. Bar Practical Guidance 3 (Duties of Competence and Diligence) (“The duty of competence requires more than the mere detection and elimination of false AI-generated results.”). A competent and diligent attorney must also read and reason. The California Court of Appeal put it well: “To state the obvious, it is a fundamental duty of attorneys to read the legal authorities they cite in appellate briefs or any other court filings to determine that the authorities stand for the propositions for which they are cited.” Noland, 114 Cal. App. 5th at 445 (original emphasis). This exercise of professional judgment “cannot be delegated to generative AI and remains the lawyer’s responsibility at all times.” Cal. Bar Practical Guidance 3 (Duties of Competence and Diligence). That is, there is no substitute for “critical attorney analysis fostered by traditional research and writing.” Id.

The duties of competence and diligence also apply, of course, to attorneys who do not use generative AI and do not suspect its use in the drafts they review. And the substance of the duties is the same. “Simply stated, no brief, pleading, motion, or any other paper filed in any court should contain any citations—whether provided by generative AI or any other source—that the attorney responsible for submitting the pleading has not personally read and verified.” Noland, 114 Cal. App. 5th at 430 (original emphasis)."

(Emphases in original)

I fully understand why the panel says that in the present case. The lawyers here hired law students who hadn't yet passed the Bar to draft their briefs, and the AI-generated results included both hallucinated (i.e., fabricated) cases as well as numerous claims about real cases for which those cases did not stand. The panel wanted to make it clear -- and rightly so -- that one's duty is not only to cite-check cases to make sure that those cases in fact exist, but that they also stand for the proposition for which they are cited in the brief.

Fair enough. That's indeed one's obligation, and it was clearly violated here.

But there's a difference, in my view, between one's responsibility in this regard, on the one hand, and the particularized personal duty that exists to ensure that this responsibility is satisfied.

Let's just reiterate for a second the money quote from today's opinion:

"No brief, pleading, motion, or any other paper filed in any court should contain any citations—whether provided by generative AI or any other source—that the attorney responsible for submitting the pleading has not personally read and verified."

Now, I've signed a lot of briefs in my day. I'm sure so have you. The ones I wrote myself, yep, I surely personally read and verified every single citation. I'm the one that put them in, and they indeed stood for the propositions for which I cited them.

There are other briefs, by contrast, where I trusted the process. Here's my admission:

I have not, in fact, personally read every single case cited anywhere in every section of every single brief I've ever signed. Important cases? Definitely read them. Ones we (or the other side) discusses at some length, or relies heavily on? For sure.

But every single case? Even for common background propositions, or orthogonal points, or positions I already know to be unambiguously true?

No. I forthrightly admit I haven't always personally read those. I may instead have trusted that those cases say what they say (e.g., that conclusions of law are reviewed de novo on appeal) because (1) I'm confident that's right, and (2) I have faith that the lower level person who cited the case looked it up themselves and made sure that's indeed what the case holds.

Now, if my trust is misplaced, and the case doesn't exist (or stand for the cited proposition), and I fully and completely responsible for that deficiency? Yes. One hundred percent. My bad. I am the one who should -- and will -- take the hit. Without complaint. It's my obligation to make sure that the cases that are cited in a brief that I sign say what I say they say.

But it seems overly strict -- and affirmatively pernicious -- to also require the signatory attorney to personally read and verify every single case that's cited. That's excessive, in my view. It increases the expense to the client -- significantly -- and unnecessarily so. You don't need a $1200/hour partner to personally read and Shepardize every single case, no matter how minor or uncontroversial. Someone has to read them, of course. And the accuracy of those citations remains the responsibility of whomever signs the brief. But responsibility can include delegation. I'm responsible for everything my underlings do. But that doesn't mean I have to personally replicate their work. That's not only inefficient, but is also detrimental to the monetary interests of my client in cost-effective lawyering.

Are there some lawyers out there who personally read and analyzing every single case, without even a single exception, that's cited in their brief? (And, presumably, Shepardize them as well to make sure they remain good law.) Surely yes. I'm confident they exist.

But I'll nonetheless wager that the majority of lawyers -- and, to be clear, I mean the majority of good, intelligent, and ethical lawyers -- do not make that their uniform practice. They do not, in fact, make sure to personally read in every matter every single case cited in a brief. They instead, on occasion, delegate. So long as such lawyers act reasonably and understand that they remain fully responsible for everything in the brief, such a practice is not per se wrong. You're not obligated in every matter to read personally every single case.

That's my controversial statement for the day, in any event.

Just know that the Ninth Circuit thinks I'm wrong. They've now said that it's your duty as a signator to personally read and analyze every single case in the brief.

So, right or wrong, as of today, that's something you gotta do. FYI.

U.S. v. $1,106,775 in Currency (9th Cir. - June 2, 2026)

This is a very professorial opinion by Judge Bress. It's got a lot of detailed introductory material regarding the discovery process in federal civil forfeiture cases, and that material is both relevant and very helpful in assessing the opinion. Very well done and well written.

The en banc court concludes that the claimant's special interrogatory responses here sufficiently evasive and unclear to justify the dismissal of his claim. That seems right to me. Dismissal is a harsh remedy. The claimant submitted multiple responses and amendments. Were those responses lacking in a lot of detail? Sure. But they were sufficient to move the case forward. Their deficiencies, and detail, could be obtained in subsequent discovery proceedings; e.g., document requests and depositions. There was no reason or justification for dismissing his claim in its entirety. Especially, as Judge Sanchez notes, while the claimant had a somewhat powerful motion to suppress pending that, if successful, might have gutted the claims of the United States to the money.

All that's well and good, and I appreciated reading the cogent opinions.

On remand, I think that the claimant here -- Oak Porcelli -- had better go all-in on his motion to suppress. Mr. Porcelli was caught on I-80 near Reno with $1,106,775 in cash in his vehicle contained in vacuum-sealed plastic bags. Mr. Porcelli, who says he's a movie producer, claims that the cash was his, and came not from drugs, but essentially from "cash [that] never went into a bank or financial institution account" as payment for his "work[] in the movie industry for 15 years, from ~1995–2010."

Uh, yeah. There are probably lots of small-time players in the movie industry who've accumulated $1.1 million in cash over the course of their career and drive it around in vacuum-sealed bags in their vehicle. Personally, I've got around $1.3 million in my Subaru right now, all neatly packed away in smell-proof bags. Just in case I need around that much in an unexpected emergency.

Unless the motion to suppress is granted, I think that the claimant here is going to have a difficult time convincing a jury that he ended up with the cash the way he said he did.

And might even face a subsequent perjury charge.

So word to the wise:

Be careful. And, if you can, win that motion to suppress. Because I think there's a lot riding on it.

People v. Five Guys (Cal. Supreme Ct. - June 1, 2026)

Were there four different published opinions from the California Supreme Court today, involving five different individuals sentenced to death? Yes. Yes there were.

Did I read each of those opinions in their entirety? Yes. I did.

Were the underlying facts of those opinions incredibly depressing, particularly when read one after the other? Oh yes. Most definitely.

Did those opinions collectively span over 600 pages? Did it take me until nearly 7:00 p.m. to finish them all? Yes and yes.

At this point in the evening, do I feel like regaling anyone with the gruesome details of those murders, and the intimate details what some innocent people, including but not limited to multiple children, were forced to endure in the final, most horrifying moments of their lives?

No, I do not.

Suffice it to say that the published opinions today were collectively amongst the most depressing ever assembled on a single day.

People v. Bankston (Cal. Supreme Ct. - June 1, 2026)

The California Supreme Court may have approved it in prior cases, but (unsurprisingly) holds today that under the Racial Justice Act, using the "Bengal Tiger story" in closing argument isn't permissible.

(I always thought that "story" was silly anyway.)