Photo of Steven Boranian

We have spilled a lot of blog ink on Federal Rule of Evidence 702 recently, so it was nice to see a case from our home state of California driving home the importance of following the rules when it comes to expert opinions.  California has a reputation for allowing expert opinions into evidence more permissively than under the Federal Rules, and that reputation is probably well deserved. 

There are, however, rules—and the plaintiff in McDonald v. Zargaryan, No. B329565, 2025 Cal. App. LEXIS 850 (Cal. Ct. App. Dec. 22, 2025), learned the hard way that there can be consequences to playing fast and loose.  In McDonald, the plaintiff claimed injuries to his hip and leg, and later to his neck and groin.  It did not, however, seem to slow him down much:  He continued to snowboard and rollerblade, but nonetheless pursued litigation and disclosed 30 experts under California’s rules requiring the exchange of expert information if any party demands it. 

This is where it gets weird.  Sixteen months after disclosing experts, and one week before trial, the plaintiff went to a new doctor, who recommended spine surgery.  As the California Court of Appeal would later describe it, “Until then, no one had proposed spine surgery [and] . . . [s]pine surgery had not been an issue in the case.”  Id. at *2.  The doctor, moreover, had a “professional relationship” with the plaintiff’s lawyer, and when asked whether his lawyer had referred him to the doctor, the plaintiff replied, “I don’t recall, but possibly.  Maybe.  I think so, before the trial.”  Id. at *3.  Hmm.

Despite these curious circumstances, the trial court allowed the (very) late disclosure of the doctor as an expert and denied a motion to exclude the testimony, so long as the plaintiff made the expert available for a deposition, which occurred.  The expert testified, and a jury returned a substantial verdict for the plaintiff.  Id. at *4-*5.

Allowing the testimony was an abuse of discretion, and the Court of Appeal reversed.  As the court explained,

The goal [of expert witness disclosure] is to avoid surprise at trial.  Surprise at trial is unfair.  It also is inefficient.  Surprise at trial is unfair because ambushes, while effective in warfare, are disfavored in court.  For legal disputes, California has replaced free-for-all trial by combat with rules of professionalism and fair play.  Surprise at trial is inefficient because, if both sides know exactly what evidence the trial will produce, they have a better chance of agreeing in advance on the true value of the case. 

Id. at *5-*6 (emphasis in original, citations omitted).  Here, the plaintiff bent the rules beyond the breaking point.  He did not seek the trial court’s permission to add a new expert, which was required, and he had no reasonable justification for the delay.  Trial counsel submitted his own declaration attempting to explain, but the declaration was just a legal brief and was “worthless as a piece of evidence.”  Id. at *10.

There can be valid justifications for late expert designation—sudden unavailability of an expert through death, illness, incapacitation, and “other serious and uncontrollable events.”  Id. at *10.  We can think of other possible reasons too, such as an unexpected change in the case or the plaintiff’s condition.  But none of that occurred here.  Not even close. 

Editor’s note: This post was revised on January 9, 2026, to remove a reference to this opinion being unpublished. The Court of Appeal certified the opinion for publication.

Photo of Eric Alexander

Perhaps driven by fear of retribution for saying what you really think, an indirect method of communication has gained some popularity on the social media platforms of late.  It goes like this:  1) a historical fact or spin on one is presented, such as on a past military conflict or a criminal conviction; and 2) there is a sentence at the end saying something like “This is not a post about [the subject matter discussed directly].”  Readers who think they know what the post was really about can feel special, whereas others may simply enjoy the bedlam that is the comments section on just about any social media post.  Still others may wish the author had opted for directness.  Our post today on this non-social (or asocial) media site concerns a case with claims and counterclaims that, on first blush, seem far from the Blog’s bailiwick.  First impressions, however, can be misleading.

In Eli Lilly & Co. v. Premier Weight Loss of Ind., LLC, No. 1:25-cv-00664-TWP-TAB, 2025 U.S. Dist. LEXIS 268138 (S.D. Ind. Dec. 31, 2025) (“PWL”), manufacturer A sued quasi-manufacturer B over trademark infringement and false advertising under the Lanham Act and its Indiana state law cousin.  Quasi-manufacturer B brought a counterclaim against manufacturer A for defamation based on statements the manufacturer had made to news outlets about its allegations in the lawsuit.  To resolve manufacturer A’s motion to dismiss, the PWL court had to delve into FDA issues that we find interesting and relevant to issues facing drug companies these days.  The connection, perhaps predictable from the names of the parties, is that the quasi-manufacturer was repackaging and selling the manufacturer’s prescription weight loss and diabetes medications.  (For some other posts on the intersection of the Lanham Act and the FDCA, try here, here, here, and here.)  This is not the increasingly common situation where a compounding pharmacy does its thing, because the defendant was not a licensed pharmacy, compounding or otherwise.  Instead, the defendant allegedly took the plaintiff’s “factory-sealed single-dose autoinjector pens containing 0.5 mL of tirzepatide fluid in various strengths,” repackaged the drug into “lower and/or different doses” in “third-party insulin syringes,” and sold those to patients with unapproved labeling and package inserts.  Id. at *3-4.  Before plaintiff sued over the alleged trademark infringement and false advertising, it shared the proposed complaint with a few local news outlets, at least one of which also allegedly received a statement saying, “We will continue to take action to stop these illegal actors and urgently call on regulators and law enforcement to do the same.”  Id. at *4-5.  The allegedly defamatory statements underpinning the counterclaim were:

1. Lilly stated that PWL is an “illegal actor” such that “regulators and law enforcement” should take action against it;

2. Lilly’s Complaint states that PWL is putting its patients’ lives at risk; and

3. Lilly’s Complaint states that PWL breaks apart or cracks open Lilly’s autoinjector pens.

Id. at *7 (internal parentheticals omitted).  Because truthfulness is a complete defense to defamation, by bringing this counterclaim, the quasi-manufacturer opened itself up to an early test of whether its conduct was illegal.

The PWL court first gave some background on the Indiana standards for defamation and the truthfulness defense, and then addressed the counterclaimant’s argument that it was premature to rule on a defense in the context of a motion to dismiss.  We will skip over that stuff to get to the substance of PWL.  Taking the allegedly defamatory third statement first, the court determined that its “gist” or “sting”—terms from the homey Indiana defamation caselaw—was true because the admitted repacking of a liquid drug from an autoinjector pen into an insulin syringe was “synonymous” with saying the counterclaimant “breaks apart” or “cracks open” the pens to extract the drug.  Id. at *13.  So, the counterclaim was dismissed as to alleged defamation based on that statement.  The second statement required little analysis because the pleadings shed insufficient light on whether the repackaging activities put patients’ lives at risk, such as by compromising sterility.  Id. at *11 & 23-24.  Accordingly, the motion to dismiss was denied on that issue.

The first statement engendered more analysis.  The counterclaimant argued that the statement could not be true unless its conduct violated a criminal statute.  The court rejected this based on Seventh Circuit law interpreting “illegal conduct” to encompass non-criminal acts.  Id. at *14.  Of course, the plaintiff’s claims did allege violations of the Lanham Act.  While evaluating on the “truthfulness” standard instead of the burden of proof applicable to the claims, the court found that “[d]ispensing the contents of Lilly’s medicines into third-party syringes and changing the packaging, labeling, and dosages contained therein, plausibly violates both the language and purpose of the Lanham Act.”  Id. at *19.  The PWL court specifically noted the importance to a trademark holder of preserving the quality control standards for the medications it manufactures.  Id, at *18.  Similar considerations informed the court’s related finding that “PWL’s conduct violates the FDCA,” a finding stated with any plausibility qualifier.  Id. at *20.  The court noted that FDA approval “constitutes approval of the product’s design, testing, intended use, manufacturing methods, performance standards, and labeling [and is] specific to the product,” and held that the counterclaimant’s repackaged products were not covered by the NDA approvals for medications and, thus, needed their own.  Id. at *22-23 (citations omitted).  Those products had not been approved, so their sale violated 21 U.S.C. § 355(a).  So, the first statement was also truthful in calling the counterclaimant an “illegal actor” based on an FDCA violation.  Like we said, this post is not really about trademarks or defamation.  Of course, by filing the counterclaim, the quasi manufacturer invited this ruling, which will make its defense of the manufacturer’s claims that much harder.  Being hacks for the companies that develop the medications and other innovative medical products in the first place, we are not too troubled by this.

The last part of the PWL decision involved the plaintiff’s argument that its statements, specifically the second one that survived the first part of the motion to dismiss, were subject to the litigation privilege applicable to “all relevant statements made in the course of a judicial proceeding, regardless of the truth or motive behind those statements.”  Id. at *25 (citation omitted).  As far as we can tell, this is a new one for the Blog.  We have had our own cases where we challenged scurrilous statements in plaintiff’s filings and where Rule 11 sanctions for unsupported positions in plaintiff’s filings were at issue.  We have also written about proceedings to go after plaintiffs’ experts or journal authors for trade libel.  But not this particular animal.  In PWL, it came down to a timing issue.  Passing around a complaint before it is filed, and presumably other statements made at that point about what a not-yet-filed lawsuit will allege, does not implicate the litigation privilege because the statements are not “made in the course of a judicial proceeding.”  Id. at *26-28.  This made us think about other targets of medical product manufacturer efforts to protect against misstatements about their (trademarked) products:  plaintiff lawyer websites and ads.  Just because the lawyers bring one suit or many suits at some point does not mean they can retroactively cloak their pre-suit statements in any litigation privilege.  This part of the post is actually about the blight of plaintiff lawyer websites and ads.  We are certainly not troubled by limiting protection for those potential vehicles for false advertising.

Photo of Michelle Yeary

Today we are talking about the decision in Govea v. Medtronic, Inc., 2025 WL 3467214 (C.D. Cal. Nov. 26, 2025). Plaintiff claimed the case was about off-label promotion. But thanks to the court taking judicial notice of PMA supplements, it’s mainly a case about on-label use and a plaintiff who waited far too long to sue.

The device at issue is an implantable stimulator used to regulate incontinence. It is prescribed when patients do not tolerate more conservative treatments and medications. The process starts with a two-week trial run where only a lead is implanted and the generator is worn outside the body. After a successful trial, the device is fully implanted. Plaintiff was prescribed the device to treat urinary incontinence. Her trial lasted only one day because the lead wire stuck to her bandage. Her physician recommended she go forward with the implant anyway and she did. Plaintiff alleges that after the implantation in 2011, she received no relief, but instead her condition worsened and she experienced painful shocks. She started asking to have the device removed in 2016. It was eventually explanted in 2018, but her surgeon did not completely remove the lead. Plaintiff claims she was unaware that the lead remained until it was revealed by x-ray in January 2024. Id. at *3. Plaintiff filed suit in December 2024 bringing claims for misrepresentation, breach of express warranty, manufacturing defect, and negligence.

At the core of all of plaintiff’s claims are allegations that the manufacturer promoted the device for an off-label use. Plaintiff claimed the device was only approved for fecal incontinence, not urinary. However, the court granted defendant’s request to take judicial notice of multiple PMA supplements and approvals which showed FDA approval of the device for both forms of incontinence. Because urinary incontinence was approved by PMA, “arguing that [the device] is inappropriate for use in that context is, in effect second-guessing the PMA process.” Id. at *9. So, plaintiff’s misrepresentation claims based on that allegation are preempted.

However, the device’s labeling states it is contraindicated in patients who “have not demonstrated an appropriate response to test stimulation.” Id. Plaintiff alleged that her device failed prior to completion of the test stimulation, that she informed defendant’s representative of that fact, and the representative recommended permanent implantation anyway. Id. at *10. Finding that was a sufficient allegation of promotion for an unapproved use, the misrepresentation claims based on that allegation were not preempted. Likewise, the same allegations were enough for the court to find plaintiff’s breach of express warranty claims not preempted. Id. at *12.

On her manufacturing defect claim, plaintiff failed to allege any violation of a specific FDA requirement for the device. She didn’t even try to rely on CGMPs. Rather, plaintiff relied on allegations that either the device had been improperly implanted or that it must be defective because the wire migrated and could not be fully removed. The former has nothing to do with the manufacture of the device and the latter is res ipsa loquitur—with the Ninth Circuit has rejected. Id. at *11. Therefore, plaintiff’s manufacturing defect claim was expressly preempted.

The court’s final preemption analysis was on the negligence claim. First, the court found that there is no state law duty to refrain from off-label promotion. The prohibition on off-label promotion is solely a function of the FDCA. So, to the extent plaintiff’s negligence claim was based on alleging such a duty, it was an impliedly preempted private attempt to enforce the FDCA. The only negligence claim that survived preemption was based on allegations of misrepresentations made during the off-label promotion. So, the claim has to be premised on false statements, not whether the statements were on or off label.   

While some of plaintiff’s claims survived preemption, for any to survive the statute of limitations challenge, plaintiff would have to significantly re-write or back off several of her current allegations.

According to the complaint, plaintiff experienced pain which she attributed to the device the entire time it was implanted. Further, she alleges she never received any relief of her symptoms from the device, but rather her incontinence worsened.  Id. at *17. She also alleges that she continued to experience pain at the incision cite following explant. What the complaint fails to allege is that plaintiff took any steps to follow up with any physician following the explant surgery, or indeed “that she took any other actions that would constitute reasonable diligence.” Id. at *16. She relies exclusively on learning in 2024 that the wire was still implanted as the point in time when she “discovered” her injury.  

Plaintiff’s injury didn’t just sneak up on her in 2024. According to her own account, it announced itself early and often. Yet plaintiff waited six years after explant to file suit. The court was unmoved by arguments that she didn’t connect the dots sooner. When a plaintiff alleges immediate and ongoing pain following a medical procedure, the law generally expects some curiosity. At least enough to ask, “Should I look into this?” The statute of limitations doesn’t pause indefinitely while a plaintiff hopes the answer will change. Being on notice doesn’t require knowing every legal theory or scientific detail. It requires awareness of an injury and a possible connection to its cause. Plaintiff had both—and still waited.

Preemption plus time-bar is a powerful one-two punch. Either is often enough; together, they end the fight early.

Photo of Bexis

This “just desserts” story caught our eyes earlier this year – a hot-shot expert witness, on artificial intelligence, no less, got caught with his own hand in the AI cookie jar.  As a result, his credibility was destroyed, and his testimony was excluded.  The litigation leading to Kohls v. Ellison, 2025 WL 66514 (D. Minn. Jan. 10, 2025), concerned a Minnesota anti-deepfake statute.  The plaintiffs were political operatives claiming a First Amendment right to create deep fakes of candidates they opposed.  Id. at *1.  The defendant hired a California-based professor to testify “about artificial intelligence (“AI”), deepfakes, and the dangers of deepfakes to free speech and democracy.”  Id.

The AI expert, however, used AI himself in preparing his material and “included fabricated material in his declaration.”  Id.  Specifically, the would-be expert “admitted that his declaration inadvertently included citations to two non-existent academic articles, and incorrectly cited the authors of a third article.”  Id.  AI had provided “fake citations to academic articles, which [the expert] failed to verify before including them in his declaration.”  Id.  The state sought to submit a belated amendment removing the fictitious citations, but the court was having none of it.  Id.

The AI expert’s AI-based report was excluded in its entirety. 

[T]he Court cannot accept false statements − innocent or not − in an expert’s declaration submitted under penalty of perjury.  Accordingly, given that the [expert] Declaration’s errors undermine its competence and credibility, the Court will exclude consideration of [that] expert testimony

Id. at *5.  “The irony. . . . a credentialed expert on the dangers of AI and misinformation, has fallen victim to the siren call of relying too heavily on AI − in a case that revolves around the dangers of AI, no less.”  Id. at *3.  Moreover, the expert had committed a cardinal, but very common, sin of a litigation-engaged expert.  He had not lived up to his usual professional standards in reaching his paid opinions:

It is particularly troubling to the Court that [the expert] typically validates citations with a reference software when he writes academic articles but did not do so when submitting the . . . Declaration. . . .  One would expect that greater attention would be paid to a document submitted under penalty of perjury than academic articles.

Id.  The expert “abdicate[d his] independent judgment and critical thinking skills in favor of ready-made, AI-generated answers.”  Id. at *4.

Even though counsel that hired this AI-dependent expert professed no knowledge of the AI-generated falsehoods in the expert’s declaration, that did not excuse them.  Rule 11 “imposes a ‘personal, nondelegable responsibility’ to ‘validate the truth and legal reasonableness of the papers filed’ in an action.”  Id. (citation and quotation marks omitted).  In this context, attorneys have an obligation “to ask their witnesses whether they have used AI in drafting their declarations and what they have done to verify any AI-generated content.”  Id.

The court ultimately excluded the AI-generated report because the false citations had destroyed the expert’s credibility and required “steep” sanctions:

[The expert’s] citation to fake, AI-generated sources in his declaration . . . shatters his credibility with this Court.  At a minimum, expert testimony is supposed to be reliable.  Fed. R. Evid. 702.  More fundamentally, signing a declaration under penalty of perjury is not a mere formality. . . .  The Court should be able to trust the indicia of truthfulness that declarations made under penalty of perjury carry, but that trust was broken here.

Moreover, citing to fake sources imposes many harms, including wasting the opposing party’s time and money, the Court’s time and resources, and reputational harms to the legal system. . . .  Courts therefore do not, and should not, make allowances for a party who cites to fake, nonexistent, misleading authorities − particularly in a document submitted under penalty of perjury.  The consequences of citing fake, AI-generated sources for attorneys and litigants are steep.  Those consequences should be no different for an expert offering testimony to assist the Court under penalty of perjury.

Id. at *4-5 (citations and quotation marks omitted).

We think that the court reached the right result in Kohls, but for the wrong reason.  The problem with AI-generated expert testimony is not limited to AI hallucinations.  Instead, it’s deeper and goes to the concept of “expertise” itself.  Who is the expert?  Is it the person who signs the report and is proffered as an expert, or is it whatever AI program the expert used?  It is well established that one expert cannot simply “parrot” the opinions of another.  We’ve written several posts that make this point.  Why should it be any different when the expert blindly parrots something that a black-box AI program spits out, rather than some other expert? 

With that question in mind, we decided to look for other decisions that have addressed experts who used AI to create their submissions.  We think that asking the right questions led to the right answer in In re Celsius Network LLC, 655 B.R. 301 (Bankr. S.D.N.Y. 2023).  Celsius Network was a bankruptcy case applying Rule 702.  The report in question, however, was not written by the expert who signed it.  Rather, it was written by AI, and for that reason it was excluded.

The [expert] Report was not written by [the expert]. Although [he] directed and guided its creation, the 172-page Report, which was generated within 72 hours, was written by artificial intelligence at the instruction of [the expert].  By his own testimony, a comprehensive human-authored report would have taken over 1,000 hours to complete.  In fact, it took [the expert] longer to read [the] report than to generate it.  The Court therefore separately evaluates the [expert] Report. . . .  [T]he Court finds that the . . . Report is unreliable and fails to meet the standard for admission.

Id. at 308.  This AI-generated expert report could not be reliable.

  • “In preparing the report, [the expert] did not review the underlying source material for any sources cited, nor does he know what his team did (or did not do) to review and summarize those materials.”
  • “There were no standards controlling the operation of the artificial intelligence that generated the Report.”
  • “The Report contained numerous errors, ranging from duplicated paragraphs to mistakes in its description of [relevant parameters].”
  • “The [expert] Report was not the product of reliable or peer-reviewed principles and methods.”

Id. at 308.  Thus, Celsius Network determined “that the Report does not meet the standard set forth under Rule 702.”  Id. at 309.

In an earlier case, an expert offered analysis of data that he had fed into a set of algorithms and “click[ed] ‘Go.’”  In re Marriott International, Inc., Customer Data Security Breach Litigation, 602 F. Supp.3d 767, 787 (D. Md. 2022).  That was not enough to be admissible under Rule 702.

Algorithms are not omniscient, omnipotent, or infallible.  They are nothing more than a systematic method of performing some particular process from a beginning to an end.  If improperly programmed, if the analytical steps incorporated within them are erroneous or incomplete, or if they are not tested to confirm their output is the product of a system or process capable of producing accurate results (a condition precedent to their admissibility), then the results they generate cannot be shown to be relevant, reliable, helpful to the fact finder, or to fit the circumstances of the particular case in which they are used. . . .  [The expert’s] willingness to rely on his own untested conclusion that his model could reliably be applied to the facts of this case is insufficient to meet the requirements of Rule 702.

Id. (footnote omitted).

A similar state-law case, Matter of Weber, 220 N.Y.S.3d 620 (N.Y. Sur. 2024), is from a state trial court.  It’s not entirely clear that the damages opinions excluded in Weber were even those of a qualified expert, the decision treated them as such and found them “inherently unreliable.”  Id. at 633.  They had been generated by an AI program (CoPilot).  The Weber expert simply parroted whatever the AI program generated:

Despite his reliance on artificial intelligence, [the expert] could not recall what input or prompt he used to assist him. . . .  He also could not state what sources [AI] relied upon and could not explain any details about how [the AI] works or how it arrives at a given output.  There was no testimony on whether these [AI] calculations considered any fund fees or tax implications.

Id.  The would-be expert nonetheless claimed that AI use was “generally accepted” in the relevant field.  Id. at 634 (New York state courts follow Frye).

The court had “no objective understanding as to how [the AI program] works,” and thus tried it out itself.  The program gave three different answers to what should have been a simple mathematical calculation – and none of those matched the supposed expert’s number.  Id. at 633 “[T]he fact there are variations at all calls into question the reliability and accuracy of [AI] to generate evidence to be relied upon in a court proceeding.”  Id.  Interestingly, when asked “are your calculations reliable enough for use in court,” the program responded that, standing alone, it was probably not ready for legal prime time.

[The AI] responded with “[w]hen it comes to legal matters, any calculations or data need to meet strict standards. I can provide accurate info, but it should always be verified by experts and accompanied by professional evaluations before being used in court. . . .  ”  It would seem that even [the program] itself self-checks and relies on human oversight and analysis.  It is clear from these responses that the developers of the [AI] program recognize the need for its supervision by a trained human operator to verify the accuracy of the submitted information as well as the output.

Id. at 634.  To prevent “garbage in, garbage out . . . a user of . . . artificial intelligence software must be trained or have knowledge of the appropriate inputs to ensure the most accurate results.”  Id. at 634 n.25.

Weber thus rejected the testimony, citing “due process issues” that “arise when decisions are made by a software program, rather than by, or at the direction of a [human].”  Id. at 634.

[T]he record is devoid of any evidence as to the reliability of [the AI program] in general, let alone as it relates to how it was applied here.  Without more, the Court cannot blindly accept as accurate, calculations which are performed by artificial intelligence.

Id.  Weber made several “finding” with respect to AI:

  • AI is “any technology that uses machine learning, natural language processing, or any other computational mechanism to simulate human intelligence, including . . . evidence creation or analysis, and legal research.”
  • “‘Generative A.I.’ [i]s artificial intelligence that is capable of generating new content (such as images or text) in response to a submitted prompt (such as a query).”
  • “[P]rior to evidence being introduced which has been generated by an artificial intelligence product or system, counsel has an affirmative duty to disclose the use of artificial intelligence.”
  • AI generated evidence “should properly be subject to a Frye hearing prior to its admission.”

Id. at 635.

Concord Music Group, Inc. v. Anthropic PBC, 2025 WL 1482734 (Mag. N.D. Cal. May 23, 2025), is another instance of an expert exposed by an AI hallucination – “a citation to an article that did not exist and whose purported authors had never worked together.”  Id. at *3.  The court considered the infraction “serious,” but not as “grave as it first appeared.”  Id.

[Proponent’s] counsel protests that this was “an honest citation mistake” but admits that Claude.ai was used to “properly format” at least three citations and, in doing so, generated a fictitious article name with inaccurate authors (who have never worked together) for the citation at issue.  That is a plain and simple AI hallucination.

Id. (citation omitted).  However, “the underlying article exists, was properly linked to and was located by a human being using Google search.”  Id.  For that reason, Concord did not view the situation as one where “attorneys and experts have abdicated their independent judgment and critical thinking skills in favor of ready-made, AI-generated answers.”  Id. (indirectly quoting Kohls).  Still the existence of the hallucination was fishy enough that the relevant paragraph from the expert report was stricken:

It is not clear how such an error − including a complete change in article title − could have escaped correction during manual cite-check by a human being. . . .  [The court’s] Civil Standing Order requires a certification “that lead trial counsel has personally verified the content’s accuracy.” Neither the certification nor verification has occurred here.

Id.  Further, as in Kohls “this issue undermines the overall credibility of [the expert’s] written declaration, a factor in the Court’s conclusion.”  Id.  Cf. Shoraka v. Bank of Am., N.A., 2023 WL 8709700, at *3 (C.D. Cal. Dec. 1, 2023) (excluding non-AI expert report that “consist[ed] almost entirely of paragraphs . . . simply copied and pasted from online sources”).

On the other hand, we have Ferlito v. Harbor Freight Tools USA, Inc., 2025 WL 1181699 (E.D.N.Y. April 23, 2025).  The plaintiff’s expert, lacking formal credentials, claimed considerable practical experience.  Among other reasons, the defendant sought to exclude his report because “after completing the report, he entered a query into ChatGPT about the best way to secure a hammer head to a handle, which produced a response consistent with his expert opinion.”  Id. at *1.  Ferlito denied exclusion because the expert had only used AI “after he had written his report to confirm his findings” – findings initially “based on his decades of experience.”  Id. at *4.  The expert “professed to being ‘quite amazed’ that the ‘ChatGPT search confirmed what [he] had already opined’” and claimed, “that he did not rely on ChatGPT.”  Id.  Taking that testimony at face value, Ferlito allowed the expert opinions:

There is no indication that [the expert] used ChatGPT to generate a report with false authority or that his use of AI would render his testimony less reliable.  Accordingly, the Court finds no issue with [the expert’s] use of ChatGPT in this instance.

Id.  Ferlito is not necessarily inconsistent with the previous decisions because of the expert’s denial that he had used AI to generate the actual report.

Considering this precedent, while it does seem that some courts are being distracted, in addressing Rule 702 issues by AI’s propensity for hallucinations, most of them do understand the more basic issue with expert use of AI – that the opinions are no longer those of the experts themselves.  Rather, when experts use AI to generate their reports, they have reduced themselves to “parrot” status, blindly reciting whatever the AI program generates.  As such, AI generated expert reports should not be admissible without some means of validating the workings of the AI algorithms themselves, which we understand is not possible in most (if not all) large-language AI models.

Photo of Bexis

As 2025 came to an end, we presented our loyal readers with our annual review of our ten worst decisions of the past year and our ten best decisions of the past year.

Now, in the new year, as we do each year, we’re pleased to announce that four (we hope) of your bloggers – Bexis, Steven Boranian, Stephen McConnell, and Lisa Baird – will be presenting a free 90-minute CLE webinar on “The Good, the Bad and the Ugly: The Best and Worst Drug/Medical Device and Vaccine Decisions of 2025” on Wednesday, January 14th at 12 p.m. EST to provide further insight and analysis on these cases.

This program is presumptively approved for 1.5 CLE credits in California, Connecticut, Illinois, New Jersey, New York, Pennsylvania, Texas and West Virginia. Applications for CLE credit will be filed in Colorado, Delaware, Florida, Georgia, Ohio, and Virginia. Attendees who are licensed in other jurisdictions will receive a uniform certificate of attendance, but Reed Smith only provides credit for the states listed. Please allow 4-6 weeks after the program to receive a certificate of attendance.

FOR VIEWERS OF RECORDED ON-DEMAND PROGRAMS: To receive CLE credit, you will need to notify Learning & Development CLE Attendance  once you have viewed the recorded program on-demand. 

**Please note – CLE credit for on-demand viewing is only available in California, Connecticut, Illinois, New Jersey, New York, Pennsylvania, Texas, and West Virginia. Credit availability expires two years from the date of the live program.

The program is free and open to anyone interested in tuning in, but you have to sign up in advance here.

Photo of Eric Hudson

Happy new year, and welcome to 2026. While we may still be pondering the meaning of auld lang syne or waxing philosophical about the new year, we’ll quickly move on  and get to work defending our clients. That’s what we do as defense hacks, and kudos to all of you for doing it so well.

We’ve written many times about plaintiffs who try (and fail) to plead injury by alleging hypothetical risks, speculative future harm, or buyer’s remorse untethered to actual loss. Today’s dismissal of a putative class action from the Northern District of California is a new year’s reminder that Article III and statutory standing remain stubbornly real requirements.  Druzgalski v. CVS Health Corp., 2025 U.S. Dist. LEXIS 265766 (C.D. Cal. Dec. 23, 2025).

Continue Reading New Year, Same Old Standing Problems
Photo of Bexis

Today’s guest post is by Reed Smith‘s Jamie Lanphear. Like Bexis, she follows tech issues as they apply to product liability litigation. In this post she discusses a pro-plaintiff piece of legislation recently introduced in Congress that would overturn the current majority rule that electronic data is not considered a “product” for purposes of strict liability, and impose product status on such data nationwide. As always our guest posters deserve 100% of the credit (and any blame) for their writings.

**********

Taking a page from the EU playbook, Senators Dick Durbin and Josh Hawley recently introduced the AI LEAD Act, a bill that would define AI systems as “products” and establish a federal product liability framework for such systems.  The structure is strikingly reminiscent of the EU’s new Product Liability Directive (“PLD”), which we previously unpacked at length here, here, and here.  Unlike the EU (aside from Ireland), however, the United States has a common-law system and a history of considering product liability to be a creature of state law.  Federal legislation to create uniform, substantive product liability principles for the entire country appears unprecedented, although an attempt forty some years ago was filibustered to death.

Like the attempt in the 1980s, it is unlikely the bill will pass in its current form (read it and you’ll understand why), but its introduction still matters.  It signals a policy tilt toward treating AI as a product, an issue U.S. courts have been wrestling with.

Before diving in, a word on tone.  The bill’s findings assert that AI systems, while promising, have already caused harm, and cite tragic incidents involving teenagers who allegedly died after “being exploited” by AI chatbots.  S. 2937, 2025, p. 2.  That is likely a nod to ongoing AI chatbot litigation—where no court has yet adjudicated liability.  Building a federal framework on such “findings,” without established liability, illustrates how current sentiment could shape future law—even if this bill never becomes one.

Key Provisions of the AI LEAD Act

The bill lifts familiar product liability doctrines into an AI-specific statute, then tweaks them in ways that matter.  The devil, as always, is in those tweaks.

  • First, causes of action. The bill would create four routes to liability: negligent design, negligent failure to warn, breach of express warranty, and strict liability for a “defective condition [that is] unreasonably dangerous.”  Conspicuously absent is a standalone manufacturing defect claim.  And unlike most state court regimes that parse strict liability by defect type, the bill would package strict liability into a single “defective condition” bucket, raising the obvious question of what counts as “defect” in a software context.
  • Second, noncompliance as defect. Under the bill, noncompliance with applicable safety statutes or regulations would deem a product defective with respect to the risks those rules target.  Compliance, by contrast, would be only evidence—it would not preclude a finding of defect.  This biased provision resembles how the new EU PLD treats noncompliance with safety regulations, though the LEAD Act is more aggressive in its approach, establishing defect rather than merely presuming it in the face of noncompliance.
  • Third, nontraditional harms. “Harm” would include not only personal injury and property damage but also “financial or reputational injury” and “distortion of a person’s behavior that would be highly offensive to a reasonable person.”  S. 2937, Sec. 3(8).  That expansion raises more questions than answers.  What exactly is “behavioral distortion”?  How is it shown?  And how would reputational injury be defined within the context of AI products?
  • Fourth, circumstantial proof of defect. Courts could infer defect from circumstantial evidence where the harm is of a kind that “ordinarily” results from a product defect and is not solely due to non-defect causes.  That is a familiar evidentiary concept in product cases involving shattered glass and exploding widgets, but translating “ordinarily” to AI, where baseline failure modes, expected outputs, and user modifications are still being defined, will be harder.  What “ordinarily” happens with a large model depends on the training corpus, the guardrails, the deployment environment, and the prompt.  In other words, results can vary widely, so it’s hard to say that any particular outcome is ordinary.  That may explain why “ordinarily” would expand liability beyond the usual state-law “malfunction theory” formulation, which requires plaintiffs to exclude reasonable secondary causes. 
  • Fifth, liability for deployers. The law would extend liability to deployers, not just developers.  “Deployers”—those who use or operate AI systems for themselves or others—could be liable as developers if they substantially modify the system (i.e., make unauthorized or unanticipated changes that alter the system’s purpose, function, or intended use) or if they intentionally misuse it contrary to intended use and proximately cause harm.  Separately, if the developer is insolvent, beyond the reach of the court, or otherwise unavailable, a deployer could be held liable to the same extent as the developer, with indemnity back against the developer if feasible.  That tracks the EU trend of extending exposure across the supply chain.
  • Sixth, a federal cause of action and limited preemption. The bill would create a federal cause of action that could be brought by the U.S. Attorney General, state AGs, individuals, or classes.  It would allow injunctive relief, damages, restitution, and the recovery of reasonable attorneys’ fees and costs.  On preemption, the bill would supersede state law only where there is a conflict, while expressly allowing states to go further.  That is not a clean sweep of state law; it is a floor with room for states to set a higher ceiling.
  • Seventh, a foreign-developer registration hook. Foreign developers would need a designated U.S. agent for service of process before making AI systems available in the U.S., with a public registry maintained by DOJ and injunctive enforcement for noncompliance.

The Bigger Picture: Software is Marching Toward “Product”

The AI LEAD Act fits a global trend of treating software and AI as products subject to strict liability.  The EU’s rebooted PLD makes this explicit.  This bill points in the same direction and, in places, pushes harder.  That matters because, as the Blog discussed in a previous post, U.S. courts traditionally treated software as a service, which often kept strict liability theories off the table.  Recent decisions, however, have nudged in the other direction, allowing product liability claims to proceed against software and AI systems.  Bexis just finished a law review article on this subject.  A federal statute that codifies AI as a “product” would accelerate that shift, harmonize some rules, and upend others.

Conclusion: What’s Next

While unlikely to pass as written, the AI LEAD Act is further evidence that AI and software are entering a new phase in the world of product liability law.  The bill reflects a growing interest in regulating AI through a product liability lens.  For companies developing or deploying AI, the practical takeaway at this stage is simple: keep watching.  Whether or not the AI LEAD Act advances, the center of gravity is moving toward treating at least some AI functionality like a product.

Photo of Michelle Yeary

Plaintiffs sometimes treat an MDL like a long layover—stretch their legs, grab a coffee, and assume that once they board the flight back to their home court, the airport rules no longer apply. Surprise! The TSA of civil procedure has a long memory, and your boarding pass still has the MDL stamp on it. Procedural orders entered in an MDL simply don’t evaporate when a case is remanded or transferred for trial. They follow the case home. And if they’re ignored, trial courts can—and do—dismiss cases for failure to comply.

Of all the things plaintiffs’ counsel need to track, there is one item that really should not fall through the cracks—whether their clients are still alive. This is not gallows humor. It’s basic federal civil procedure. Rule 25 requires a motion to substitute a proper party-plaintiff within ninety days of the filing of a suggestion of death. But Rule 25 does not set a deadline for the filing of a suggestion of death. Unfortunately, in MDLs with hundreds or thousands of cases it becomes dangerously easy for individual plaintiffs to become names on a spreadsheet rather than people whose status needs monitoring. Then months (or even years) go by before counsel and the court learn that the MDL inventory is full of deceased plaintiffs and procedurally defective cases. And since that can distort settlement metrics, bellwether pools, and case valuations, some MDL courts enter additional requirements to force plaintiffs’ counsel to keep track of their clients.

Such a pretrial order (PTO) was entered in the Bair Hugger MDL. In addition to Rule 25, it required plaintiffs’ counsel to file suggestions of death within ninety days of the death of a plaintiff or risk dismissal with prejudice. The MDL court, in fact, dismissed several plaintiffs for failing to file timely suggestions of death. See post here. But the plaintiff in Robinson v. 3M Company, 2025 U.S. Dist. LEXIS 263427 (M.D. Fla. Dec. 22, 2025), apparently thought that because the MDL judge didn’t personally dismiss their case before transferring it out of the MDL, the slate must be clean. That’s a bold strategy, Cotton.

Plaintiff Robinson passed away on June 4, 2025. So, to comply with the MDL PTO, plaintiff needed to file a suggestion of death by September 2, 2025. They did not, despite learning of the death on August 22nd.  Instead, defendants filed a suggestion of death on September 16th upon independently learning of the death and then moved to dismiss the case. Id. at *3-4.

Plaintiff’s main argument against dismissal was that the MDL PTO did not apply to remanded cases. However, that ignores that the Robinson case was not remanded. Because it was directly filed in the MDL, it was “transferred” to Florida for trial, not “remanded.” The PTO itself explicitly stated that it applied to all directly filed cases and the order transferring the case explicitly incorporated “Selected Orders” filed in the MDL, including the order regarding suggestions of death. Id. at *5-7.

Therefore, a suggestion of death should have been filed within 90 days of plaintiff’s death—and not, as plaintiff argued within 90 days of plaintiff’s counsel learning of the death. “The strict deadline is intended to ensure the timely progression of litigation and to prevent delays arising from counsel’s lack of diligence or communication.” Id. at *7. Plaintiff offered no excuse for not having a procedure in place to communicate with their clients at least every ninety days which would have enabled them to comply with the PTO. Nor was plaintiff entitled to ninety days from the filing of the suggestion of death to file a motion to substitute. Having failed to comply with the PTO’s threshold requirement, defendant was entitled to move to dismiss.  Id. at *8-9.

But here is where the transferor court cut the tardy plaintiff a little slack. The MDL PTO called for a dismissal with prejudice, but Eleventh Circuit precedent favored a without prejudice dismissal in this situation. So that is what the court ordered. Id. at *10-11.  

While the distinction between remand and transfer played a role here, nobody should assume that in either scenario the case reboots to Level One. The receiving trial court inherits the case as it stands and courts routinely hold that MDL orders remain binding post-remand/transfer. Think of it less like a reset and more like a baton pass. MDL courts issue real orders with real consequences. And when your case goes home, those orders pack their bags and come with you. So don’t be shocked when a trial court enforces them. The only surprising thing would be if it didn’t.

Photo of Bexis

A year ago, Bexis thought this would be the last blogpost he would ever write, and that he would retire along with 2025.  Not gonna happen – at least not yet, thanks to Reed Smith making him an offer he couldn’t refuse.  Bexis will be stepping back from some things, but not from the Blog.  That will continue.

With the future of the Blog assured for at least another year, we turn to Bexis’ favorite holiday task, compiling our annual list of the top ten best prescription medical product liability litigation decisions.  For a judicial decision to be eligible for our annual top (or bottom) ten lists, it must:  (1) involve a prescription medical product, or an OTC drug, and (2) involve product liability claims, that is, the losing plaintiff must have claimed that there was something wrong with that product.

So, before we turn to the best of the best for 2025, we’re going to pay passing homage to those non-prescription medical product or non-product liability decisions in 2025 that we think will significantly impact our sandbox in years to come.  Non-prescription medical product decisions of significance in 2025 are:  Gustafson v. Springfield, Inc., 333 A.3d 651 (Pa. 2025) (nothing in the Tenth Amendment precluded Congress from enacting federal tort reform) (here); Smith v. Terumo BCT, Inc., ___ P.3d ___, 2025 WL 3029699 (Colo. App. Oct. 30, 2025) (rejecting non-injury medical monitoring as a basis for tort recovery under Colorado law) (here); EcoFactor, Inc. v. Google LLC, 137 F.4th 1333 (Fed. Cir. 2025) (en banc) (excellent expert exclusion decision applying amended Rule 702 in a patent case; failure to exclude expert was abuse of discretion) (here); Engilis v. Monsanto Co., 151 F.4th 1040 (9th Cir. 2025) (another good Rule 702 decision from previously problem-child circuit) (here); Messerli v. AW Distributing, Inc., 153 F.4th 1077 (10th Cir. 2025) (strongly reaffirming in pari delecto as an independent basis for dismissal) (here); Pable v. Chicago Transit Authority, 145 F.4th 712 (7th Cir. 2025) (sanctions against plaintiff spoliating ediscovery affirmed) (here); In re Subpoena No. 25-1431-014, ___ F. Supp.3d ___, 2025 WL 3252648 (E.D. Pa. Nov. 21, 2025) (defending off-label use of FDA approved products from government arguments that mere manufacturer knowledge of such use renders such products “unapproved new drugs”) (here); United States v. Novo Nordisk, Inc., 2025 WL 1696881 (W.D. Wash. June 17, 2025) (FCA governmental plaintiffs have to provide relevant discovery, too) (here); LLT Management, LLC v. Emory, 2025 WL 438100 (E.D. Va. Feb. 7, 2025) (talc defendant could pursue trade libel suit against p-side experts who allegedly deliberately lied in authoring a litigation-driven junk science medical journal article) (here).

But the potential impact of those decisions doesn’t in any way detract from the significance of our top ten decisions, as these ten are directly applicable to the arguments we make in defense of our prescription medical product clients.  This year’s Drug & Device Law Blog top ten decisions of the year reaffirmed vigorous vetting of would-be expert witnesses under recently amended Rule 702.  They applied preemption – our most powerful defense.  They rejected public nuisance claims against prescription medical products.  They enforced early vetting of MDL plaintiffs.  So we are chomping at the bit to get going with our 2025 top ten prescription medical product liability decisions.

  1. This entry is not from the Reed Smith side of the Blog.  In re Zantac (Ranitidine) Litigation, 342 A.3d 1131 (Del. 2025).  When plaintiffs in the Zantac MDL began losing major issues right and left (see our Zantac Chronicles (2020+10, 2020+18, 2022+4, 2024+19) plaintiffs bailed out of that federal forum in droves, even though they had promised not to.  Many of these litigation tourist refugees wound up in Delaware state court, where several of the defendants were incorporated.  There, they sought and received an expert do-over under what purported to be the Delaware Rules of Evidence.  The result was, in a word, horrible, our worst trial court opinion of last year (2024-5).  We’re not going to go through everything wrong with that opinion again.  You can read about it here.  But before the misdeeds of a single trial judge could do permanent damage to Delaware’s reputation, the Delaware Supreme Court stepped in, and in a rare move accepted an interlocutory appeal.  The outcome?  Our number one best decision of 2025.  The result was even bigger than this litigation, and the Delaware Zantac litigation involved some 75,000 plaintiffs.  Beyond that, the Delaware Supreme Court expressly adopted the standards of amended Federal Rule 702, including the 2023 amendment – completely repairing Delaware law in the expert arena, which is critical given the number of our clients incorporated there.  First, the court rejected the trial court’s overall expert admissibility standard for improperly favoring and presuming admissibility.  As in federal court, Delaware judges are “gatekeepers” and must examine proffered expert testimony critically.  Second, the Supreme Court specifically agreed that the bases of expert opinions are for the court, not the jury, to evaluate under Rule 702.  And with that, it reversed the trial court, found multiple abuses of discretion, and ordered the ten (count ‘em 10) causation witness for these 75,000 plaintiffs all be excluded for a variety of improper methodological lapses.  The worst of these lapses was relying on studies that, instead of involving the product the litigation was about, concerned the alleged toxic agent, in either occupational or dietary exposures that dwarfed anything that these plaintiffs could have received.  In so doing, it explicitly adopted the reasoning of the federal Zantac MDL’s exclusion decision (2022+4).  Since nine of the ten experts had improperly based their opinions on occupational/dietary studies, and the last one could not offer a complete opinion without the others, the trial court’s decision admitting plaintiffs’ expert testimony was reversed – the opinions were inadmissible under Rule 702, properly applied.  And recently, a postscript:  no expert-related mulligans; summary judgment was proper, and the only remaining question is which plaintiffs are bound by the summary judgment order.  See In re Zantac (Ranitidine) Litigation, 2025 WL 3459343 (Del. Super. Dec. 1, 2025).  We (some of us) zealously advocated the reasoning in Zantac here.
  2. This entry is solely from the Holland & Knight side of the Blog.  Eastern Maine Medical Center v. Walgreen Co., 331 A.3d 380 (Me. 2025).  A good winning streak is a thing to behold, all the more if it starts with a series of setbacks.  Just about every feel good sports movie seems to follow that formula.  In the early years of opioid litigation, the plaintiffs seem to hold all the cards, pushing staggering potential liability for dozens of companies based primarily on the public nuisance theory of liability.  But with decisions by the Supreme Courts of Oklahoma (2021+1) and Ohio (2024+3) that rejected public nuisance as a prescription medical product liability litigation theory, it looked like public nuisance was in sharp retreat.  This year, a third state high court came out against the proposed application of public nuisance in opioid litigation.  In Eastern Maine, however, the plaintiffs were not local governments but hospitals claiming the opioid epidemic harmed them by increasing their provision of unreimbursed care.  Maine had adopted public nuisance (also called “common nuisance” Down East) under its common law and enacted a statute authorizing damages for private plaintiffs.  Neither mattered in the end.  Despite their 538-page complaint, plaintiffs in Eastern Maine could not “show an infringement of private rights resulting in special legal injury different in kind as well as degree from that suffered by others.”  Their attempt to glom onto “the broad public injury resulting from increased opioid misuse” did not get them past pleadings for public nuisance, negligence, fraud, or conspiracy.  We had hoped that as goes Maine, so goes the nation, but the favorable public nuisance appellate trend ended with Fourth Circuit’s mess purporting to apply West Virginia law (2025-1).  For us, however, three state supreme court decisions (and the Third Restatement of Torts) outweigh a single circuit court’s improper purported Erie prediction of state law.  A few of us exalted Eastern Maine here.
  3. Sprafka v. Medical Device Business Services, Inc., 139 F.4th 656 (8th Cir. 2025).  From the beginning, we’ve believed that the December, 2023 amendments toughening Fed. R. Evid. 702 − and calling out “liberal” admissibility decisions as “incorrect” – could be a game changer in favor of defendants.  Given the lag time in litigation, 2024 is the first year with significant amended Rule 702 appellate decisions.  Sprafka, which affirmed an excellent district court decision (2024+11), is the best in the drug/device space, so it’s number three on our annual nice list.  The Eighth Circuit has been a Rule 702 problem child (2021-1) for quite some time, but Sprafka cited the amended rule in affirming the exclusion of a notorious p-side frequent flyer “expert” from testifying in a medical device case.  While we would have preferred a frank recognition that the amended rule superseded that circuit’s past bad decisions, Sprafka gave us the next best thing:  a published decision discussing the impact of the 2023 amendments and a detailed affirmance of the trial court’s exclusion order.  The opinions at issue flunked Rule 702 because they were solely litigation-driven, had inadequate support that failed to establish any failure rate, were not supported by reliable data, and failed to offer any causal alternative design/modifications to the device.  The purported opinion that this design had a higher failure rate than other similar devices was based on anecdotes that were flatly contradicted by relevant registry data.  The expert’s defect opinion was thus “properly disregarded” due to unreliable methodology.  We saluted Sprafka here.
  4. In re Gardasil Products Liability Litigation, 151 F.4th 178 (4th Cir. 2025).  What did the plaintiffs in the first-ever MDL involving a Vaccine Act-protected vaccine try to do?  They tried to destroy the Vaccine Act itself.  Fortunately, they failed miserably.  They didn’t have any worthwhile claims that were not preempted by the Act, so they claimed that the Act was unconstitutional because the process by which every new vaccine (including Gardasil) was added to the list of covered vaccines for the last 40-years supposedly violated the constitution’s Presentment Clause.  They failed, and in this Gardasil decision that failure was affirmed.  Because Congress explicitly created the Act’s listing  process and set defined criteria for its use, it did not grant improper  executive branch authority to make statutory amendments.  But that’s not all.  Gardasil affirmed a second, unrelated pro-defense ruling also being appealed.  This ruling is less important, but still significant.  On their way through the Vaccine Court, special masters held that several plaintiffs’ claims were not timely filed (no surprise since MDL solicitation inevitably dredges up stale and otherwise meritless claims).  In the MDL, these plaintiffs unsuccessfully tried to collaterally attack the masters’ decision (2024+14).  Dismissal of these plaintiffs’ cases based on the collateral estoppel effect of the special master decisions affirmed.  The Vaccine Act vests timeliness decisions in the special masters, and plaintiffs never properly appealed to the Federal Court of Claims.  Having taken no appeal, plaintiffs cannot avoid the special master decisions in subsequent civil litigation.  For rejecting both of these MDL-inspired attacks on the Vaccine Act, the Fourth Circuit’s Gardasil decision ranks high among our best cases of the year.  We greeted Gardasil here.
  5. Wilson v. CooperSurgical, Inc., ___ F. Supp.3d ___, 2025 WL 2606086 (S.D. Ill. Sept. 9, 2025).  We rank Wilson as our best trial court decision of the year because it correctly recognized that failure-to-report claims do not state a claim under Illinois law.  Thus, the plaintiff’s claims against the manufacturer of a pre-market approved contraceptive device were entirely preempted.  Supposedly “parallel” failure-to-report claims have emerged as a preemption dodge of choice for device, and even drug, plaintiffs faced with extinction through preemption.  Wilson is also noteworthy for the judge’s complete about face on preemption once the plaintiff’s case moved beyond the pleadings.  We criticized the prior bad decision as our sixth worst case of 2023.  It’s only fair that we reward this favorable turn of events.  2025 will go down as the year that anti-choice plaintiffs’ attacks on yet another form of contraception got clobbered on summary judgment, and Wilson heads that list.  Illinois failure-to-report law has been murky because of a couple of early and adventurously bad federal court decisions.  But a 2017 published Illinois appellate decision rejected reporting-based claims.  In an extensive discussion, Wilson analyzed the conflicting precedents and determined that “there is no Illinois requirement that parallels the federal requirement to report to the FDA.”  There being no state-law parallel duty, the claim was expressly preempted for being “different from or in addition to” FDA requirements.  The reporting claim was also impliedly preempted under the Buckman rationale that private plaintiffs have no ability to enforce purported FDCA violations.  Like the fraud-on-the-FDA claim at issue in Buckman, plaintiff’s reporting claim sought to second-guess the completeness of the defendant’s submissions to the FDA.  The reporting claims being the linchpin of all the Wilson plaintiff’s allegations (typical of litigation over this product), negligence and gross negligence also bit the dust.  We welcomed Wilson here.
  6. In re Valsartan, Losartan, & Irbesartan Products Liability Litigation, 2025 WL 1024048 (D.N.J. April 7, 2025) (“Valsartan 2025”).  The Valsartan MDL, during its first five years, was notable for several bad decisions (2024-8, 2021-4, 2020-10), culminating in the jaw-dropping certification of no fewer than three nationwide class actions (2023-1).  Fortunately for the defense, in 2025, there was a new sheriff in town – and boy did it matter with Valsartan 2025 making our “best of” list.  The plaintiffs first tried to argue that the new judge was somehow constrained by “law of the case” and had to follow a previous ruling finding an expert’s testimony admissible in the context of class certification.  Nope, the merits are wholly different, being in the “world of proof and evidence.”  Then the Valsartan 2025 plaintiffs lost badly on said proof and evidence.  The class representative in the third-party payor class action could not prove anything because it could not prove cancer causation.  Drugs that successfully treated the conditions for which they were prescribed could not be “worthless” because of a trivial level of claimed contamination by an alleged carcinogenic break-down product.  Plaintiffs could not avoid the “elephant in the room” – could this medicine, in the doses prescribed, cause cancer – by seeking only economic loss damages.  The economist that the class plaintiffs proffered was neither qualified to address cancer causation, nor did she purport to do so.  Given this “stark lack of scientific or economic basis,” Valsartan 2025 reject the expert’s opinions as virtually all ipse dixit:  assuming, rather than proving dangerous levels of contamination.  Icing on the cake was provided by the court’s favorable references to the 2023 amendments to Rule 702.  We viewed the virtues of Valsartan 2025 here.
  7. In re Gardasil Products Liability Litigation, 770 F. Supp.3d 893 (W.D.N.C. 2025).  The plaintiffs in the Gardasil MDL, being bound by the preemptive limits of the Vaccine Act, see  above #4), were stuck with warning claims.  But even putting aside the Act’s presumption that FDA-approved warnings were adequate, the Gardasil plaintiffs had another, ultimately insurmountable problem − impossibility preemption.  Plaintiffs had not, and could not, dispute the nearly 100% effectiveness of the vaccine in preventing a once common form of cancer.  Instead, they alleged that the vaccine caused either of two relatively minor conditions.  The evidence that it did so was pathetically slim – no more than a couple dozen largely unverified reports.  Thus, without even bothering with the statutory presumption, the warning claims were entirely preempted.  An FDA-approved warning can only be challenged if, under applicable FDA standards of evidence, a plaintiff has “newly acquired information” of some new or more frequent/serious medical risk.  The scraps that plaintiffs relied on didn’t come close.  Because preemption, and thus the existence or the requisite information, was a question of law, plaintiffs couldn’t rely on paid “experts” spouting ridiculous theories (although they tried).  For one thing, all of the information, sparse as it was, had been provided to the FDA.  Thus, there was nothing “new.”  Nor were case reports, published or not, sufficient evidence.  They resided at the bottom of the scientific weight scale.  Still worse were “fewer than ten” adverse event reports.  Even putting aside their completely unverified nature “these reports would be far below the background rate” of the complained-of conditions – nothing more frequent.  Medical literature that didn’t even mention the relevant conditions simply didn’t count.  Because plaintiffs had nothing approaching the necessary FDA evidentiary standard for unilateral label changes, all plaintiffs’ warning claims were impliedly preempted as “impossible” under the governing FDCA framework.  To allow such claims “would effectively make the regulatory standard meaningless.”  Another great Gardasil decision in an MDL the plaintiffs probably regret ever starting.  We gloried in Gardasil here.
  8. In re Depo-Provera (Depot Medroxyprogesterone Acetate) Products Liability Litigation, 2025 WL 1618995 (N.D. Fla. May 12, 2025).  As many of our readers know, a new federal rule, Fed. R. Civ. P. 16.1, pertaining specifically to MDLs, went into effect at the beginning of this month.  We are admittedly skeptical about the new rule, because it didn’t directly address what to us is the overwhelming problem with MDLs – the filling of thousands of utterly unvetted claims gathered by widespread attorney solicitation.  The new rule’s one saving grace on the issue of early vetting of MDL claims was that “early exchange of information” about the basis of claims (and defenses) is a mandatory topic to be addressed early on under new Rule 16.1.  That was why we considered the new rule “better than nothing,” but nothing could give it a run for its money.  On the upside, Depo-Provera promises more, much more, if the terms of this order become the norm for MDLs, which is why this order makes our top ten list.  It represents an excellent model for MDL early vetting with teeth under Rule 16.1, even though it preceded the effective date of the rule by several months.  Defendants in other MDLs should seek similar orders.  It imposed pleading requirements.  Every complaint in the MDL must have (or be amended to have) a statement of “requisite physical injury” from a specified list and also specific allegations of causation by the plaintiff’s use of a “requisite product,” again from a list.  Enforcement was both quick (two days for a deficiency finding and seven days for an order to show cause) and effective (deficient complaints cannot be filed).  If Depo-Provera becomes the model for Rule 16.1 “exchange” requirements, then the rule will have indeed accomplished something significantly useful, and we will happily eat our words.  We delightedly presented Depo-Provera here.
  9. This entry is not from the Butler Snow side of the Blog.  Silverstein v. CoolsSculpting Zeltiq Aesthetics, Inc., 229 N.Y.S.3d 44 (N.Y. App. Div. 2025).  Silverstein is a relatively rare published state-court opinion flatly rejecting an increasingly common, and thoroughly obnoxious, plaintiff argument – that product liability defendants have duties to warn about the risk of products made by others.  For one thing, consider innovator liability, which at its core is precisely that kind of theory.  Asbestos cases rejecting the so-called “bare metal defense” are another example.  Silverstein rejects the same arguments in a one-off medical device case, and did so in the best possible procedural posture, unanimously reversing the denial of summary judgment (an appealable order in New York).  Use of the defendant’s product in cosmetic surgery was commonly followed by application of ice packs to the affected area.  Unfortunately, extreme cold, like extreme heat, can burn the skin.  However, the defendant that was sued did not make or sell icepacks, and its device by itself did not have a burn risk.  Relying on one of the bad asbestos cases, the trial court in Silverstein denied summary judgment, finding the frequent combined use of the two products supported a duty to warn.  Reversing, the appellate court held that ice packs were not essential to the defendant’s device function.  Nor did the defendant’s warnings recommend use of ice packs.   Finally, the defendant didn’t make ice packs and thus received no economic benefit from their sale.  The general rule confining the duty to warn to products the manufacturer made (and profited from) thus applied.  And there’s more.  Because the device required a prescription, the learned intermediary rule also applied.  To a physician who regularly used this sort of device, the danger of burns from ice packs was obvious – “basic medical knowledge.”  There was thus no need to warn in the first place of this risk and even if there were, no causation, since the doctor using the device was well aware of that risk.  We salivated over Silverstein here.
  10. In re Glucagon-Like Peptide-1 Receptor Agonists Products Liability Litigation, 2025 WL 2396801 (E.D. Pa. Aug. 15, 2025).  With this year’s overall emphasis on enforcement of amended Rule 702, it is only fitting that another favorable MDL-wide expert ruling should make the top ten, while several other excellent decisions receive honorable mentions.  The GLP-1 MDL is essentially a strike MDL; because the drugs are widely used, plaintiffs went looking for some claimed injury – didn’t matter what − to tag it with.  The MDL judge seems to have similar thoughts, since from the beginning she limited MDL proceedings to “cross-cutting” issues, one of which was medical: “whether a physician can reliably diagnose a patient with gastroparesis without performing a gastric emptying study.”  While that sounds obscure, 95% of the claims in the MDL involve this alleged injury.  Plaintiffs, of course, viewed any objective criteria as anathema.  Fortunately for the right side of the “v.,” this test is precisely what consensus medical practice required.  Thus, plaintiffs’ experts’ unconstrained “differential diagnoses” (actually etiologies) failed Rule 702’s methodology test.  The aforementioned medical consensus was virtually unanimous, and their experts stretched the only two even arguably supportive studies “further than the authors themselves were willing to go.”  Those experts also used case studies, which were simply not “an accepted method of diagnosis.”  Other litigation-created hypotheses that these experts generated were excluded as completely unsupported speculation or as constantly moving targets.  With no scientific basis for avoiding objective testing, plaintiffs claimed “regulatory estoppel” (which sounds like a new name for fraud on the FDA), but even if such a thing existed, plaintiffs had taken the statements out of context.  So any plaintiff that didn’t have the requisite objective test at the time of diagnosis is out.  A similar Rule 702 specific-test requirement (2024+7) successfully cut the Zostavax MDL down to size.  May the same thing happen here.  We were gladdened by GLP-1 here.

There you have it – our top ten prescription medical product liability litigation decisions of 2025.  We think it’s a good mix of substantive law, preemption, Rule 702, and procedural rulings, most of which were magnified by occurring in the mass tort context.  If you were responsible for any of these gems, your grateful bloggers offer our congratulations.  Otherwise, we will wait together for next year.

But we’re not done yet.  Recognizing defense wins is fun, so here is our annual list of runners up − those decisions that we consider to be the next ten most significant wins of 2025.

Honorable Mentions:  (11) Bulox v. Coopersurgical, Inc., 773 F. Supp.3d 351 (S.D. Tex. 2025) (it was a good year for arguments that failure-to-report claims in the Filshie clip litigation were preempted; this is another outstanding summary judgment decision) (here); (12) In re Gardasil Products Liability Litigation, 2025 WL 1782576 (W.D.N.C. Feb. 20, 2025) (the Gardasil MDL has become a disaster for the plaintiffs; this decision applied amended Rule 702 to give another set of lousy experts a well-deserved bum’s rush) (here); (13) Saulsby v. Amphastar Pharmaceuticals, Inc., 919 S.E.2d 727 (N.C. App. 2025) (excellent appellate affirmance of dismissal of an OTC drug plaintiff who ignored warnings; would have ranked higher if framed in terms of warning causation rather than North Carolina’s peculiar retention of “contributory negligence” as a complete defense) (here); (14) Paddock v. Novartis Pharmaceuticals Corp., 2025 WL 1908806 (D. Del. July 11, 2025) (good on snap removal, but even better for recognizing fraudulent misjoinder as grounds for removal) (here); (15) In re Taxotere (Docetaxel) Products Liability Litigation, 2025 WL 655796 (5th Cir. Feb. 28, 2025) (non-precedential affirmance of with-prejudice non pros of an MDL plaintiff who couldn’t be bothered even to effect service of process on the defendants) (here); (16) In re Valsartan, Losartan & Irbesartan Products Liability Litigation, 2025 WL 3131002 (D.N.J. Nov. 10, 2025) (first MDL bellwether plaintiff dismissed because his specific causation expert could not reliably exclude other strong and established cancer risk factors) (here); (17) Searcy v. Pfizer, Inc., __ F. Supp.3d __, 2025 WL 2713736 (M.D. Ala. Sept. 23, 2025) (because the PREP Act effectively shut down product liability claims involving COVID-19 vaccines, this plaintiff claimed the Act was unconstitutional in various ways – and lost them all) (here); (18) Thelen v. Somatics, LLC, 156 F.4th 1115 (11th Cir. 2025) (published appellate affirmance of defense trial win on adequate warnings that also governed a consumer expectation-based design claim and evidentiary rulings; ranked lower because of application of Nebraska law by a Florida court) (here); (19) Hill v. Medical Device Business Services, Inc., 2025 WL 1950300 (6th Cir. July 16, 2025) (broken device case that produced another favorable Rule 702 affirmance, but unpublished) (here); (20) Watts v. Maryland CVS Pharmacy, LLC, 142 F.4th 233 (4th Cir. 2025) (excellent published affirmance of dismissal based on 50/50 causation being inherently speculative, but involving a rare fact pattern of two simultaneous vaccinations, only one being subject to Vaccine Act preemption) (here).

But we still can’t stop ourselves from patting victorious defendants on the back.  Our 2025 collection of cases also features these near misses – another group of useful victories, listed mostly in chronological order:

Summers v. Medtronic, Inc., 2025 WL 863576 (D. Mass. March 18, 2025) (complete dismissal on PMA preemption; suffering only from “making rubble bounce” issue) (here); Warner v. Amgen, Inc., 2025 WL 490720 (D. Mass. Feb. 13, 2025) (refusing to recognize the PMA preemption dodge of “pre-approval” warning claims) (here); Arnold v. CooperSurgical, Inc., 2025 WL 622075 (S.D. Ohio Feb. 26, 2025) (another summary judgment Filshie clip preemption win, flipping an adverse Rule 12 decision and finding no viable failure-to-report liability theory in Ohio) (here); Scholl v. Walgreens Specialty Pharmacy, LLC, 2025 WL 950866 (N.D. Okla. March 28, 2025) (pharmacy that allegedly delayed filling a prescription had no state-law duty requiring it to fill any particular prescription) (here); Laucella v. Medtronic, Inc., 2025 WL 1018414 (C.D. Cal. April 4, 2025) (fraudulent joinder found where plaintiff purported to join an in-state non-treating physician who consulted with the defendant about the device, followed by PMA preemption dismissal of entire action) (here); Doyle v. Bayer Corp., 2025 WL 1666261 (W.D. Wash. June 12, 2025) (“threadbare” complaint against IUD manufacturer dismissed on multiple grounds, from preemption to pleading) (here); Aguila v. RQM+ LLC, 2025 WL 2322438 (S.D. Fla. Aug. 12, 2025) (American registered agent for FDA purposes not an agent for service of legal process on foreign manufacturer) (here); Bacher v. Boehringer Ingelheim Pharmaceuticals, Inc., 2025 WL 2463143 (Conn. Super. Aug. 22, 2025) (another horde of refugees from the Zantac MDL gets dismissed; this time on personal jurisdictional grounds) (here).

Looking to the future, we are all waiting with bated breath for the California Supreme Court’s decision in the Gilead duty-to-innovate case (2024-1), but oral argument has yet to be scheduled.  The longer it takes, the more nervous we get.  2026 should see the Second Circuit deciding the appeal from the Rule 702 dismissal in the Acetaminophen (2023+4) MDL litigation.  Also on appeal is the hideous Dressen (2024-9) PREP Act decision.  Multiple Zantac appeals (2022+4) (2021+10) (2021+18) remain pending in what has turned into a procedural morass.  We didn’t see any appeal in Bueno/Parker (2024+5) on PACER, so we assume those wins are final.  Finally, there will be no Michigan Supreme Court learned intermediary rule decision, as the litigation that produced a certification order has been dismissed.

The Supreme Court will decide whether a defendant that successfully removes a case to federal court is always at risk that, should the basis for removal later be found invalid on appeal, all subsequent activity would be nullified, essentially giving plaintiffs who lose in removed cases (and would be the only appellants) a free do-over (here).  The Eleventh Circuit will opine on whether the False Claims Act’s qui tam provision unconstitutionally vests executive power in private plaintiffs (here).

The new MDL rule, Fed. R. Civ. P. 16.1 just became effective, so in 2026 should see whether it will have the beneficial effects on multi-district litigation that we hope (see Depo-Provera, above) – especially on early vetting of claims − as opposed to the minimal effect that we fear could be the case.  Another potentially interesting new rule, proposed Fed. R. Evid. 707, regarding admissibility of machine-generated evidence, remains pending before the relevant civil rules subcommittee.

Looking back, we addressed the Delaware Zantac (2024-5) reversal above and discussed last week (2025-1), how the City of Huntington decision (2022+8) was reversed.  The “unprecedented” use of the political question doctrine in Caston (2024+13) also unfortunately did not survive appeal.  Certiorari was denied in Fosamax (2024-2), review was denied in Herzog (2024-7), and transfer was denied in O’Haver (2024+18), so those are over.  Beaver (2024+20) was affirmed by a one-paragraph order.  Beaver v. Pfizer, Inc., 2024 WL 3617570 (4th Cir. Aug. 1, 2024).

Finally, there is one piece of pending national legislation – the so-called “Lead Act” that would overturn the current majority rule that intangible electronic information is not a “product” for strict liability purposes – that bears watching.  We will have a post about that in the new year (actually, we beat that).

See you in 2026.

Photo of Steven Boranian

It is December 26, and all is quiet on the home front.  The Drug and Device Law Family exchanged gifts, took long walks, and ate the traditional herb-encrusted rib roast.  The year is nearly over, and having reached a certain age, we can say for sure that time flies when you are having fun.  We hope all our readers are enjoying the holiday season and have enjoyed our daily musings—posted every morning, almost always current, and sometimes even insightful.  Thanks for reading.

Time flew by for the plaintiffs in Miller v. Rush University Medical Center too, but the outcome for them was anything but fun.  In Miller, a federal judge in Illinois struck the plaintiffs’ Fourth Amended Complaint and denied their motion to remand, mainly on the basis that the plaintiffs were obviously trying to destroy diversity.  No. 23-cv-002210, 2025 U.S. Dist. LEXIS 258154 (N.D. Ill. Apr. 14, 2025).  In striking the amended complaint, the district court went so far as to question the plaintiffs’ motive and found that they had mischaracterized the record.

Here is what happened.  The plaintiffs sued several medical providers and medical device manufacturers and distributors following back surgery.  The plaintiffs amended their complaint three times, and the defendants moved to dismiss each time.  Before the court ruled on the last motion to dismiss, the plaintiffs sought leave to amend to file a Fourth Amended Complaint. 

This is where things started to go wrong for the plaintiffs.  To start, they represented that they had learned about new defendants, including an individual “who may share culpability with other named Defendants.”  The problem there is that the plaintiffs represented that this was a recent discovery, but in fact they had known about this individual for nearly a year.  In addition, the plaintiffs represented that the proposed amended complaint “should have no impact on the pending Motions to Dismiss” and urged the court to “freely give leave” to amend under Rule 15(a).  This was not altogether true either.  The plaintiffs disclosed, for the first time, in the Fourth Amended Complaint that the citizenship of the new defendants destroyed diversity.  This would very much “impact” the Motions to Dismiss, as it would divest the district court of jurisdiction to rule on them. 

The defendants thus moved to strike the Fourth Amended Complaint and opposed remand on the basis that, had the court known the proposed amendments would destroy diversity, it would not have granted leave to amend under Rule 15.  Instead, the court would have conducted an analysis under 28 U.S.C. § 1447(e), which governs diversity-destroying amendments.

When joinder of a nondiverse party would destroy subject matter jurisdiction, Section 1447(e) applies and gives the district court two options:  (1) deny joinder, or (2) permit joinder and remand the action to state court.  Miller, at *7.  To guide that discretion, the Seventh Circuit has adopted four factors:  (1) the plaintiff’s motive for seeking joinder, particularly whether the purpose is to defeat federal jurisdiction; (2) the timeliness of the request to amend; (3) whether the plaintiff will be significantly injured if joinder is not allowed: and (4) any other relevant equitable considerations.  Id. at *7-*8 (citing Schur v. L.A. Weight Loss Ctrs., Inc., 577 F.3d 752, 759 (7th Cir. 2009)). 

The plaintiffs’ attempt here to add nondiverse defendants failed on all counts.  First, their motive clearly was to destroy diversity.  They wrote that their recent discovery of the new defendants prompted their last motion for leave to amend, but that was not true.  They knew about the “new” defendants even before they filed their Third Amended Complaint.  Moreover, the court found that the plaintiffs were playing fast and loose with the facts.  Or, as the court put it, “Plaintiffs mischaracterize both the extent of discovery they received and the facts contained in the discovery.”  Id. at *8-*9.  Ouch.  Still further, the court found that the allegations against the new defendants were bogus, i.e., they lacked a “plausible factual basis” and a “reasonable basis in law.”  The only conclusion was that the plaintiffs’ motion was something other than a genuine desire to pursue these defendants.

Second, the timing of the motion to amend suggested a sideways motive, too.  “As a general rule, an extensive delay between removal and a motion to amend typically weighs against permitting joinder, though the rule may not apply when the plaintiff only recently learned about the non-diverse defendants’ roles.”  Id. at *16 (citing Schur, 577 F.3d at 767).  Here, the plaintiffs had the information that they relied on for almost a year.

Third, denying the joinder would not harm the plaintiffs because their claims against the diversity-destroying defendants were not viable anyway.  And fourth, equitable factors weighed against joinder because the plaintiffs’ amendments had caused delay and had deprived the defendants of a ruling on their Motion to Dismiss. 

The court therefore granted the defendants’ motion to strike the Fourth Amended Complaint, reinstated the Third Amended Complaint, and promised a ruling on the Motions to Dismiss in due course.  It is impossible to read this order and not get the sense that this court felt like it had been fooled.  The plaintiffs persuaded the court to grant them leave to file a Fourth Amended Complaint, thus disrupting the defendants’ pending motion to dismiss.  But in obtaining that dispensation, the plaintiffs were not entirely forthcoming with the court, which is putting it generously.  Their motive was evident, and the court set it right. 

Drug & Device Resources

Bexis’ Book

FDA-regulated products now account for an estimated one-fifth of overall economic activity in the U.S. They have also been the focus of a litigation explosion. This timely guide covers all aspects of litigation involving drugs, medical devices, vaccines and other FDA-regulated prescription products.
Learn more about this book