More than forty years ago, the Supreme Court observed that “[e]xperienced advocates since time beyond memory have emphasized the importance of winnowing out weaker arguments … and focusing on one central issue if possible, or at most on a few key issues.” Jones v. Barnes, 463 U.S. 745, 751–52 (1983). The principle is a simple one: the strongest briefs seldom contain the greatest number of arguments but rather exemplify restraint by selecting and developing the arguments most likely to prevail. Since then, this kernel of Supreme Court wisdom has taken root in the federal courts, especially in the Seventh Circuit, where courts have repeatedly condemned “scattershot,” “kitchen sink,” and “blunderbuss” approaches to briefing as “the antithesis of sound advocacy.” [citations omitted].

The court explained:

The reason is straightforward: weak arguments do not just fail on their own merits—they often weaken stronger ones. As Judge Kennelly of the United States District Court for the Northern District of Illinois has explained, advocates invariably operate within page limits, so “[t]he more space you use on the less meritorious points, the less you have for the stronger ones.” Hon. Matthew F. Kennelly, Over-Arguing Your Case, 40 No. 2 Litig. 41, 42 (2014). Worse still, spending time on weaker points “detract[s] from the force of your stronger and more significant points,” leaving judges to wonder “why, if those are such good arguments, the lawyers spent so much time on inconsequential points.” Id. Judge Easterbrook has captured the same concern: a brief advancing too many arguments “sends the message that counsel does not think much of any of the claims raised—or perhaps does not believe the court able to separate good arguments from bad.” Max M., 859 F.2d at 1300.

Scale AI requested permission to file an oversized brief.  Permission was granted. The court wrote:

Yet the additional space used in the opening briefs did not meaningfully expand the development of Scale’s strongest arguments or factual allegations. Instead, Scale devoted substantial portions of its briefs to arguments for which it lacked standing to pursue, while leaving several potentially dispositive theories confined to only a sentence or two, often with sparse citation to the administrative record. Had Scale concentrated its fire on the bullseye rather than peppering the entire target, the additional pages it sought—the majority of which were ultimately left unused—might have been devoted to developing the arguments that actually may have mattered.

In conclusion, the court wrote: “In the end, Scale spent too much time raising arguments that could not establish Article III standing nor render Scale eligible for award, even if resolved in its favor. At the same time, Scale spent too little time on arguments that mattered—skimping on both development and citation to, and drawing upon, the administrative record. In other words, rather than concentrating its fire on the bullseye, Scale put its efforts toward issues lying in the outer rings. Following such an approach, Scale failed to meet its high burden of establishing that NGA’s award decision was arbitrary, capricious, an abuse of discretion, or otherwise contrary to law.”

In a prior post, I wrote that sometimes government personnel use personal systems to conduct government business.  I suggested that it is not a good idea.

A Reuters article authored by Nate Raymond, Former Fauci adviser pleads guilty to conspiring to conceal records (Aug. 19, 2026), reinforces that idea:

A former adviser to infectious disease expert Anthony Fauci pleaded guilty on Tuesday to conspiring to evade public records laws and conceal government documents related to grant research funding and the COVID-19 pandemic….  The 78-year-old had been indicted in April on charges related to what prosecutors said was a scheme to thwart public records requests received by the agency beginning in April 2020 related to COVID-19 research grants.

The relevant part of the Reuters article is:  “Anticipating their communications would be sought through Freedom of Information Act requests, Morens and others agreed to correspond using Morens’ personal email account instead of his government one, prosecutors said.” [emphasis added].

The State opposed the motion, arguing that it had “spent considerable time” in “detailing” the inaccuracies. It argued that “Appellant should not now be permitted to refile the Opening Brief to address the shortcomings the State identified in its Answer Brief.”

A different ADD attorney entered an appearance and reported that the prior attorney “informed ADD that he used an artificial intelligence (‘AI’) tool in preparing the brief and [the prior attorney]  failed to sufficiently review the brief for accuracy prior to filing it.”  Later, ADD reported that prior counsel had said that he used AI to “revise” the brief. New counsel stated that the prior attorney had been removed from the case.

The new attorney “renewed the request to allow the Opening Brief to be withdrawn and requested leave for new counsel to be permitted to file a new Opening Brief.”

The Supreme Court of Montana agreed with the State:

After due consideration, we are denying the motion to withdraw the Opening Brief in this appeal. We agree with the State that Feme’s appellate counsel may address the issues in the Reply Brief. M. R. App. P. 12(3) provides, in relevant part, that reply briefs must be confined to new matter raised in the Appellee’s answer brief. In this instance, we will allow new appellate counsel greater latitude in addressing the State’s arguments, including correcting factual errors and offering more precise legal authority in support of Appellant’s arguments and we further extend the deadline for filing the brief.

The court ordered that “Appellant shall prepare, file, and serve the Reply Brief within 30 days of the date of this Order.”

Before addressing the merits of the issue raised on appeal, we discuss one issue of concern. Appellant’s brief, which was filed on July 15, 2025, contained multiple inaccurate case citations. On September 3, 2025, appellant filed a notice of errata, attempting to correct the citation irregularities. On September 9, 2025, this Court treated the notice as a motion to correct the brief and denied the motion. On September 10, 2025, appellant filed a motion to strike his brief and file a corrected brief. On September 17, 2025, this Court denied appellant’s request, noting that it was filed three days before the State’s deadline to file its brief.

On October 29, 2025, this Court issued its opinion in Mezu v. Mezu, 267 Md. App. 354, 367 (2025), noting “the problems that occur when lawyers use AI [artificial intelligence], without the diligence required, in drafting briefs.” In that case, where the brief cited multiple cases that did not exist, as well as others that did not support the proposition for which they were cited, and counsel stated at argument that he did not typically read the cases he cited to the Court, we referred the attorney to the Attorney Grievance Commission….

Here, counsel for appellant has not admitted that the mistakes in the brief were due to the use of AI, but that seems likely. In assessing our response to the errors here, we reiterate that “[t]he failure to use AI responsibly in legal research raises ethical issues and can result in sanctions when used improperly. It is unquestionably improper for an attorney to submit a brief with fake cases generated by AI.” ….

In this case, however, the brief was filed before the decision in Mezu, the inaccuracies were not excessive and did not include totally made up cases, counsel tried to correct the errors before the State filed its brief, and the State does not contend that it was harmed. Given these circumstances, we shall not, as we did in Mezu, refer counsel to the Attorney Grievance Commission. We again urge counsel, however, to use AI responsibly and check the accuracy of all citations before submitting a brief to this Court, or any other court. [emphasis added].

In assessing sanctions, some courts have looked at two factors—verification and candor.  Cite Checking to Find Hallucinated Cases Deemed Insufficient – Part II – Verification and Candor Are Expected (Aug. 17, 2026).

“This is an unreported opinion. This opinion may not be cited as precedent within the rule of stare decisis. It may be cited for persuasive value only if the citation conforms to Md. Rule 1-104(a)(2)(B).”

Plaintiffs asserted that 21 documents were protected because:

each contain “internal legislative analysis,” “draft advocacy-related communications,” and/or “internal policy analysis and strategic deliberations concerning pending legislation,” … that reveal “Veterans Guardian’s internal processes for formulating policy positions, coordinating advocacy efforts, and engaging with the legislative process … in furtherance of the Company’s rights to petition the government and to associate with others.” … Plaintiffs further assert that compelled disclosure of these documents would chill future association and petition activity, harms that a confidentiality order cannot prevent.

Id. at *1 (cleaned up).  Defendant objected:

arguing that there is no First Amendment privilege for “internal legislative analysis” or “draft advocacy documents”; that there is no First Amendment right to refuse to provide evidence in a case a plaintiff has chosen to bring; and that the First Amendment privilege is not applicable in discovery, much less in discovery governed by confidentiality order…. Defendant further submits that even if the First Amendment privilege could apply in these circumstances, plaintiffs have failed to provide  any factual basis that could support a prima facie showing that these documents are properly subject to any privilege…. Accordingly, defendant submits that plaintiffs must produce the documents at issue, or, at the very least, provide sufficient information about them to allow defendant to properly assess the asserted privilege.

Id. (cleaned up).

The court ruled after a telephone conference and in camera review:

The First Amendment includes a right to associate…. Relevant here, that right protects individuals’ and organizations’ expressive association, or association for the purpose of engaging in activities protected by the First Amendment…. When state action directly affects a group’s expressive associational activities, the state must show a compelling interest that justifies the level of the burden imposed on the groups’ expression…. Discovery requests by a government litigant represent state action that, in some circumstances, can impermissibly infringe an organization’s associational activities…. When a party in litigation invokes the First Amendment privilege to withhold otherwise-discoverable materials on that basis, the party opposing discovery must show that enforcement of the discovery request will result in consequences which objectively suggest a ‘chilling’ impact on associational rights…. This prima facie showing requires appellants to demonstrate that enforcement of the discovery requests will result in (1) harassment, membership withdrawal, or discouragement of new members, or (2) other consequences which objectively suggest an impact on, or chilling of, the members’ associational rights.

Id. (emphasis added; cleaned up). If the party invoking the privilege makes a prima facie showing of chilling, the burden shifts to the government to show a compelling need for the information.

The court held that, while the privilege typically applies to materials such as membership lists, it may extend to an organization’s internal deliberations and communications.  Id. at *3.  Further, it rejected the argument that filing suit is a categorical waiver. Id.

However, on the facts presented the court held that Plaintiffs failed to meet their burden of proof.  Specifically, it wrote that Plaintiffs have “offered no facts to suggest that disclosure of the materials would chill association rights.”  Their conclusory assertion, unsupported by evidence, failed.  “Simply saying it does not make it so….  Moreover, on its independent review of the materials in camera, the Court is not persuaded that compelling their disclosure in these circumstances will in fact chill Veterans Guardian’s associational rights.” Id. at *3.

This blog was initially posted on  Electronic Discovery Reference Model

Defendants’ sanctions motion was unsuccessful because Defendants failed to provide clear and convincing evidence that the paper documents went missing after this Plaintiff’s duty to preserve was triggered.  The case demonstrates that the trigger and loss dates for each litigant is fundamental to spoliation analysis. Terran Biosciences, Inc. v. Compass Pathfinder Limited, 2026 WL 2349408 (D. Md. Aug. 13, 2026)(Crawford, J.).

THE FACTS,COMPLAINTS, and DISCOVERY REQUEST

While one plaintiff, Dr. Thompson, was working at the University of Maryland-Baltimore (“UMB”) lab, he developed a breakthrough medical treatment. The research details were recorded in physical lab notebooks and stored in the lab.

In March 2020, a major renovation began.  The materials were supposed to be moved to the floor below.  Post-renovation cleanup was completed in late 2022, however, Dr. Thompson had resigned on June 30, 2022.

The Court wrote: “When Dr. Thompson resigned, his lab ceased to exist, and there is no evidence that the removed lab materials, including the lab notebooks, ever returned to the fifth floor…. UMB has no knowledge of what happened to the lab notebooks after Dr. Thompson’s departure in June 2022.”

Plaintiffs sued, alleging improper disclosure and use of their trade secrets by Defendants.  Terran filed suit on August 5, 2022. Dr. Thompson joined as a plaintiff on June 5, 2023. UMB joined as a plaintiff on February 6, 2024.

Compass and two individuals were named defendants.

On July 10, 2025, [defendant] Compass served on Plaintiffs a request for production of documents that sought, among other things, documents relating to Dr. Thompson’s research, specifically identifying “lab notebooks” as encompassed by the request…. Defendants deposed [plaintiff] Dr. Thompson on September 16, 2025…. At this deposition, Dr. Thompson testified that when he left UMB, he knew where he left the lab notebooks but that “nobody seems able to find them.”

EFFORTS TO FIND THE NOTEBOOKS – DISCOVERY ON DISCOVERY

A discovery dispute was deferred while UMB continued to look for the notebooks. The parties then filed a status report and reported that they had agreed, in my words, to “discovery on discovery.”[1]

Specifically, the parties represented to the Court that UMB had “agreed to produce its internal litigation hold correspondence and the positions of individuals who received the litigation hold.”[2] …

UMB had also agreed to produce a witness to testify on five topics:

(1) When UMB first became aware that Terran intended to file this lawsuit, UMB’s involvement in that effort, and UMB’s knowledge of the fact that the suit involved alleged trade secrets detailed in Professor Thompson’s lab notebooks; (2) The details of UMB’s litigation hold; (3) What steps UMB took to identify relevant physical evidence, including the lab notebooks and when it implemented its litigation hold; (4) The timing and circumstances of the loss of the lab notebooks; and (5) When UMB first learned that the lab notebooks were missing, when it told Terran, and what steps it took next.

… The letter advised that after Defendants received this discovery, they would “confer with UMB on the scope of sanctions Defendants will seek based on the loss of the lab notebooks.” …

All of those steps were accomplished and the parties proposed a formal briefing schedule, which the Court granted.

COURT PROCEEDINGS

Initially, the Magistrate Judge rejected two preliminary challenges mounted by Plaintiffs.  Those were a challenge alleging untimeliness under Judge Grimm’s decision in Goodman v. Praxair, 632 F. Supp. 2d 494 (D. Md. 2009), and failure to follow the Judge’s protocol for resolving disputes.  The Court’s exhaustive analysis of timeliness is worth reading, id. at *3, but not the focus of this post.[3]

THE SPOLIATION STANDARD

Turning to the merits, the Court found sanctions were “unwarranted.”  Because the notebooks were not ESI, Fed.R.Civ.P. 37(e) did not apply and the Court held that the party moving for sanctions must show:

(1) [T]he party having control over the evidence had an obligation to preserve it when it was destroyed or altered; (2) the destruction or loss was accompanied by a ‘culpable state of mind;’ and (3) the evidence that was destroyed or altered was ‘relevant’ to the claims or defenses of the party that sought the discovery of the spoliated evidence, to the extent that a reasonable factfinder could conclude that the lost evidence would have supported the claims or defenses of the party that sought it.

The Court relied on its “inherent power” to control the judicial process, noting that inherent power is limited to curative sanctions and the movant bears the burden of providing “clear and convincing” evidence.

TRIGGER DATE

The first element involves deciding when the duty to preserve was triggered and when the loss or destruction occurred. Defendants contended that Plaintiffs’ duty arose in April 2022, months before litigation began, and long before there was any contention of loss. Plaintiffs replied that there was no duty in April 2022 and the lab notebooks most likely were lost in July 2022, before the duty was triggered.

The Facts Relating to the Trigger Date

The Court set the scene:  “There is no dispute that Terran filed this lawsuit on August 5, 2022, notified UMB of the filing on August 8, and that approximately one week later, on August 16, UMB issued its litigation hold.”  While UMB did not enter the case until February 6, 2024, it is a plaintiff and a plaintiff’s duty is triggered when it decides to pursue a claim.

In support of their asserted April 2022 trigger date, Defendants marshalled several facts.  The Complaints alleged: “Plaintiffs first suspected Defendants’ disclosure of, and misuse of, the Psilocybin Trade Secrets in approximately April 2022, in the course of reviewing Compass’s published patent application filings.”  Further, the Defendants relied on Plaintiffs’ privilege log which showed communications with outside counsel regarding “litigation, IP advice, and licensing advice.”[4]

Plaintiffs responded that UMB did not anticipate that it would later join the case.  They asserted that while Terran may have anticipated litigation in April 2022, that knowledge could not be imputed to UMB.  And, they asserted that the logged communications with counsel were on a licensing agreement with Terran, not on litigation against the Defendants.

The Court’s Holding Re: Trigger Date

The Court held that Defendants had failed to establish by clear and convincing evidence that Plaintiffs reasonably anticipated litigation as early as April 2022. It added:

Even assuming that Plaintiffs first suspected misuse of the trade secrets in April 2022, the mere suspicion of certain facts giving rise to a dispute demonstrates only that there was the “possibility of litigation,” not that Plaintiffs should have reasonably anticipated litigation at that time…. The Complaint does not allege that Plaintiffs understood in April that Compass had misappropriated their trade secrets, nor does it allege that Plaintiffs made known their suspicions of misappropriation to Compass. On the contrary, the allegation that Plaintiffs “first suspected misuse” in April suggests that this was their initial reaction to information contained in Compass’s patent filings—a far cry from an affirmative decision to sue, or even a direct threat of litigation from Terran to Compass that would have triggered a duty to preserve.  [emphasis in original].

The Court found the privilege log entries to be ambiguous. “At most, they establish that Plaintiffs communicated with private, outside attorneys regarding licensing and intellectual property issues and ‘litigation’ of an unspecified nature. That the subject emails covered multiple topics other than litigation tends to support the conclusion that litigation (of whatever nature) was not the parties’ sole, or even primary, concern at that time.”

The Court identified another gap in proof: “Nor have Defendants set forth any clear and convincing evidence establishing at what time after April 2022 but before August 5, 2022 [the date suit was filed], Plaintiffs, including then non-party UMB, may have reasonably anticipated litigation sufficient to trigger a duty to preserve.”

The Court concluded:

Accordingly, Defendants have failed to establish by clear and convincing evidence that Plaintiffs should have reasonably anticipated litigation in April 2022, thus triggering their duty to preserve evidence at that time. Rather, the evidence submitted supports the Court’s findings that Terran commenced litigation on August 5, 2022, triggering its own duty to preserve relevant evidence, and UMB’s duty to preserve arose at the latest on August 8, 2022, with UMB’s receipt of correspondence from Terran advising UMB of the lawsuit and seeking preservation of relevant evidence.[5]

The Court added: “Having determined when the duty to preserve arose, the Court now considers when UMB lost the notebooks, as the timing of when the evidence was lost in relation to when the duty to preserve arose, is essential to the spoliation analysis.”

DATE OF LOSS OF THE NOTEBOOKS

The Court wrote: “After engaging in discovery to determine the lab notebooks’ whereabouts, neither Plaintiffs nor Defendants can state with any precision when the lab notebooks were lost or destroyed. The evidence shows that the last time the lab notebooks were observed was in June 2022…. There is no concrete evidence regarding what happened to the notebooks after that time.”

Plaintiffs asserted that the “most likely” time of loss was July 2022. They relied on an October 1, 2025, UMB email stating that the lab notebooks were inadvertently lost during the renovations and cleanup.  Renovations began in early 2021 and continued into 2022, well after Dr. Thompson resigned.

The Court disagreed with Plaintiffs:

This evidence does not support the firm conclusion that the date of loss or destruction of the lab notebooks was in July 2022, versus a later point, like August or thereafter. Plaintiffs’ chosen timing is convenient because it would mean that the lab notebooks were lost before the Plaintiffs’ duty to preserve arose. But another plausible and equally speculative interpretation of the same October 1 email is that the lab notebooks may have been lost sometime after August 5, 2022, when this lawsuit commenced. The record is unclear. At bottom, the only conclusions that the Court can draw from the evidence presented are: (1) the lab notebooks went missing sometime after June 2022, when Dr. Thompson last saw them; (2) the lab notebooks were declared lost on October 1, 2025, when UMB advised Defendants of the results of their investigation into the whereabouts of the lab notebooks; and (3) the lab notebooks were in UMB’s possession when they were lost or destroyed. [emphasis in original].

However, the Court saw no need to decide when the notebooks went missing:

As noted above, I cannot conclude that Plaintiffs’ duty to preserve the lab notebooks arose in April 2022. The evidence only establishes that the duty arose on August 5, 2022, at the earliest. Because the lab notebooks were lost at some point after June 2022, it is entirely possible that the lab notebooks were destroyed in the intervening month before there was a duty to preserve. Given that Defendants have not carried their evidentiary burden at the first step in the analysis, the Court need not proceed to step two to determine whether UMB had a culpable state of mind. Doing so would require the Court to stack all its findings upon a hypothetical—a “what if” scenario that presumes Plaintiffs lost or destroyed the lab notebooks after August 5, 2022. Such a conditional analysis could not serve as the basis for the award of any sanctions.

In short, the movant failed to demonstrate that the lab notebooks went missing after the duty to preserve was triggered. The Court added: “Defendants’ evidence would also fail to meet the less stringent preponderance of the evidence standard, if such standard applied here.” Here, the trigger date was dispositive.

This blog was initially posted on  Electronic Discovery Reference Model.

____

[1] For more on “satellite” or “culpability” discovery, see generally Guidelines for “Discovery on Discovery” (Aug. 5, 2025); “Discovery on Discovery” Permitted (Sep. 4, 2025); “Discovery on Discovery” Denied in “It Ends With Us” Lawsuit  (Sep. 5, 2025).

[2] For more on disclosure of implementation communications, see “The Court is keenly interested in whether Defendants’ counsel issued a litigation hold.” (Feb. 19, 2026); 8 Lessons Learned – Part IV – Disclosure of Recipients of Litigation Hold Notices (May 9, 2024); Discovery of Steps Taken to Implement a Litigation Hold is Permitted (Mar. 26, 2022); When is a Litigation Hold Notice Discoverable? (Dec. 1, 2020)

[3] For more on timing, see, The “Best Time” to File a Spoliation Motion (Jul. 15, 2025); When Should a Motion for Sanctions be Filed?  (Mar. 30, 2026).

[4] A Privilege Log Can Be Admissible Evidence (Jan. 20, 2025); Privilege Log Entry Was Relevant to Inquiry Notice Where Time-Bar Was at Issue (Jul. 6, 2025);  Privilege Log Helps Defeat Summary Judgment on One Count (Jun. 24, 2025); Listing a Document on a Privilege Log Concedes That it is Relevant for Discovery (Sep. 6, 2025).

[5] Generally a Third Party Had No Duty to Preserve (Feb. 14, 2026).  UMB was not a party when Terra made its preservation request.

A court recently held that an upload to Claude waived any right to seal documents filed in court.

In Fry v. Fry, 2026 WL 2531956 (E.D. Penn. Aug. 27, 2026), the petitioner alleged that his wife had wrongfully removed their children from Germany to the U.S.  Petitioner also sought sealing of certain documents that had been filed.  They contained “personal identifiers such as social security numbers, dates of birth; financial account numbers; names of minor children; names and contact information of third-party military and medical personnel mentioned in his email correspondence and military records; prescription and medical records of Petitioner; correspondence with German officials; and details of domestic disputes between Petitioner and Respondent.”

After analyzing both the First Amendment and common law, and after in camera review, the court granted, in part, and denied, in part, the motion to seal.  It wrote in pertinent part:

With Petitioner’s renewed motion, the Court required Petitioner to submit an AI certification describing any assistance of AI that Petitioner utilized to prepare his motion…. Petitioner disclosed to the Court that he had uploaded his redacted documents to non-enterprise versions of Claude and Claude CoWork, products by third-party company Anthropic, in preparing his motion. As such, the Court finds that the full sealing of any document is not appropriate. Documents and information uploaded to Claude are processed on Anthropic’s servers, and Anthropic’s data privacy and data training pages make clear that such information is not private. See, e.g., Privacy Policy, Anthropic https://www.anthropic.com/legal/privacy (last accessed August 24, 2026); How Do You Use Personal Data in Model Training?, Anthropic, https://privacy.claude.com/en/articles/10023555-how-do-you-use-personal-data-in-model-training (last accessed August 27, 2026); see also United States v. Heppner, 820 F. Supp. 3d 292, 296 (S.D.N.Y. 2026) (holding that communications with Claude are not privileged due, in part, to Claude’s privacy policy). As such, it would be inappropriate for the Court to seal a document that has already been provided to a private, for-profit research and artificial intelligence company. [emphasis added].

“Three-strata validation” was requested and rejected in In re Class Action Settlement Administration Litigation, 2026 WL 2444936 (D.D.C. Aug. 19, 2026). The parties in that case had narrowed their disputes and their “only remaining disagreement concerns how a party producing documents in discovery must validate that its productions are substantially complete.”

Plaintiffs asked for “three-strata validation.”  Defendants objected to the second and third strata.

The court explained the “three-strata” approach as:  “validation of documents coded as non-responsive, validation of documents coded as responsive, and validation of documents excluded from the universe of ESI material subject to discovery review.”

THE PARTIES AGREED TO NULL SET TESTING

The parties agreed that there is an obligation  “to confirm that responsive documents are not erroneously coded as non-responsive and omitted from their productions.” This involves review of a “statistically valid random sample” of documents coded as non-responsive. Id. at *1.

This standard practice is commonly called “elusion testing,” review of the “discard pile,” or, review of  the “null set.”  See Deal Genius, LLC v. O2COOL, LLC, 682 F. Supp. 3d 727, 734–35 (N.D. Ill. 2023) (“explaining that elusion testing, where the producing party reviews a random sample from the ‘null set’ of documents that did not hit on search terms and documents that were deemed non-responsive, is a ‘standard quality assurance practice’ that can ‘confirm the reasonableness and proportionality’ of the production”).

VALIDATION OF DOCUMENTS CODED AS RESPONSIVE – REJECTED

Defendants argued that validation of documents coded as responsive, to find false positives, was superfluous.  They asserted that Fed.R.Civ.P. 26(g) required that they certify that their response is complete and correct.  The court wrote: “Rephrased in statistical terms, producing parties must confirm that their false-negative rate falls within acceptable limits.”  In my less elegant words, a production may include nonresponsive documents; but, it cannot be a “dump truck.”

The court explained:

Review of documents marked responsive falls outside the bounds of litigants’ Rule 26(g) obligations because such review cannot identify whether documents are missing from a production, rendering it incomplete. Confirming that documents designated responsive are truly responsive reveals only the production’s false-positive rate, a proxy measure for whether the litigant has overproduced discovery material. But the Federal Rules do not prohibit parties from overproducing documents. Accordingly, this Court will not require defendants to incur additional costs re-reviewing documents that they have previously coded as responsive and intend to produce, barring claims of privilege. [emphasis in original].

VALIDATION OF DOCUMENTS EXCLUDED FROM THE UNIVERSE – REJECTED

The second strata consists of “ documents excluded from the universe of ESI material subject to discovery review.”  In other words, Plaintiffs asked for a review of documents that Defendants had not collected and that, presumably, were  not in Defendants’ litigation review platform.

The court wrote that examination of this set of documents, the “Excluded Set,” “could reveal false-negatives: responsive materials that should be produced but were erroneously omitted from the universe of materials collected for discovery review.”

However, the court responded that Plaintiffs’ request was duplicative and made too late, long after collection.  It wrote:  “Yet the Court is puzzled why plaintiffs request review of the Excluded Set after discovery review is largely complete, given that the parties also intend to review the Excluded Set for responsive material prior to beginning that review.” [emphasis in original].

The court explained that “ESI discovery generally proceeds in two phases: document collection and merits review.”  The parties had earlier “set forth agreed upon parameters for collecting the universe of documents that will be subject to merits review.”  That earlier agreement had required statistical sampling.

Insofar as the parties have agreed to statistically validate the completeness of the set of documents collected for merits review before even beginning that review, the Court finds that further validation of the Excluded Set after merits review is largely complete would be duplicative and unnecessary.

However, Plaintiffs made a different argument—so-called “substantial completeness evaluation.”  They argued that:

[T]heir proposed validation efforts “answer different questions at different points in time,” because search term validation “tests whether agreed terms are performing as intended before review begins,” while substantial completeness validation of the Excluded Set “tests whether the entire process — including all downstream coding decisions, review workflow, and production choices — resulted in a complete production at its conclusion.”

This argument was unsuccessful because the court held that “the Excluded Set is a collection of documents excluded before merits review begins and not subject to downstream coding decisions, review workflows, and production choices.”

Plaintiffs also contended that “the prevailing practice is for parties producing discovery to validate the Excluded Set both upon its creation and after merits review….”  The court noted that the authorities are split; however, it held that “the Court will not require producing parties to incur the cost and burden to re-validate the Excluded Set at the substantial completeness stage, provided that they statistically validate that documents excluded from merits review do not contain responsive material before beginning that review.”

I would add that “substantial completeness validation” of the Excluded Set could de-rail a case and mandate a “do over.”  If a sufficient number of false negatives was found in that set—and sufficiency is always debatable–presumably Defendants would have to engage in a new collection, review, and production process.

This blog was initially posted on  Electronic Discovery Reference Model

In Campbell v. TidalHealth, Inc., 2026 WL 2162749 (D. Md. Jul. 28, 2026)(Austin, J.), the Court wrote: “Campbell denies submitting documents into any generative artificial intelligence tools but admits to using unspecified ‘available tools solely to express [her] arguments in proper legal format, the same way any litigant uses available legal resources.’” Id. at *8.

After cautioning Campell about the use of GenAI, the Court wrote:

In addition, Campbell must ensure that any use of generative artificial intelligence does not violate the stipulated order governing confidential materials…. TidalHealth expresses concern that Campbell violated the February 27, 2026 Protective Order’s prohibition on submitting discovery material to public generative artificial intelligence tools. … see ECF 21 at ¶ 8 (imposing limitations on use of generative artificial intelligence). Campbell denies doing so…. If future filings suggest improper submission of discovery materials to generative artificial intelligence tools, the Court will be more inclined to issue a show cause order and consider sanctions.

Id. at *10 (emphasis added).

These orders are standard of care and are most often entered by consent. But not all orders are created equal and they are not always fair.  I have echoed Craig Ball’s novel post on the need for equity.  Another A.I. Protective Order – Part 10 quoting Craig Ball, The AI Protective Order Double Standard | Ball in your Court (Jul. 27, 2026).

UPDATE:

“Because the parties’ proposed protective order adequately complies with the standards set forth above, the Court will enter an order based on the parties’ submission. The Court makes four minor modifications to the parties’ protective order: … and (4) to establish parameters for the use of Artificial Intelligence as it pertains to confidential documents, see infra Paragraph (F).”  Zask v.Kohl’s Corp., 2026 WL 2531856 (E.D. Wisc. Aug. 27, 2026)(emphasis added). The order states:

F. USE OF ARTIFICIAL INTELLIGENCE (“AI”)
A receiving party may use AI tools in connection with discovery materials produced in this action only if the AI tool is operated in a secure environment and is subject to enforceable contractual terms that (a) prohibit the provider from retaining, using, disclosing, selling, or using the materials to train, improve, or develop any AI system; (b) prohibit the materials from being made publicly available or accessible to any other customer or third party; and (c) require commercially reasonable administrative, technical, and physical safeguards to protect the confidentiality of the materials. Discovery materials, whether designated CONFIDENTIAL or ATTORNEYS’ EYES ONLY, shall not be entered into any publicly available or consumer-based AI platform. Any use of AI tools must remain consistent with this Order and with the requirement that discovery materials be used solely for purposes of this litigation.

An expert witness testifying in a lawsuit about liability for a Houston explosion that killed three people and destroyed roughly 200 homes used ChatGPT to write significant portions of his “expert report.” The man, who was hired by the industrial product conglomerate 3M, exposed his AI prompts publicly. They showed that he asked ChatGPT to help him “create an exceptional expert witness report defending the standard of care at 3M,” and that the report should “show how 3M is 0% at fault for the explosion at Watson Grinding.”

Mr. Koebler added:

The incident shows that artificial intelligence has made its way into courtrooms not just in AI-generated legal briefings, hallucinated cases, and adversarial “prompt injections,” but in expert witness testimonies. Court transcripts, deposition documents, and discovery records shared with 404 Media show extensive AI use in an extremely high profile case, where multiple people died and hundreds of millions of dollars in total liability are at stake in ongoing litigation about the explosion. The case also shows that the specific prompts used to create this type of expert testimony can be discoverable during a case, and that those prompts can be quite embarrassing.

The lawsuit centers on an explosion allegedly caused by a “degraded and poorly crimped” hose.  Dozens of homeowners have sued 3M.  Id.  The article reports that the expert was paid roughly $90,000.

The expert’s use of AI was discovered when plaintiff’s attorney, Will Moye, found a five-page document called “Citation Overlay” which he recognized as coming from ChatGPT.  He then demanded production of all of the prompts used to create the report.  Mr. Koebler reported:

ChatGPT created a roughly 30-page report that included the line “From a technical and standard-of-care standpoint, 3M is 0% responsible for the January 24, 2020 explosion.” This line did not make it into the final report filed with the court, because when Autenrieth later asked ChatGPT to “review this as the opposing council,” ChatGPT determined that writing “‘0% responsible’ is an easy target” for a lawyer to poke holes in, and is one of several “phrases [that] let opposing counsel paint you as an advocate rather than an expert.”

Mr. Moye is quoted as stating that the expert “acknowledged [at trial] the prompts he put in were biased toward 3M to help 3M win the case […] it’s really egregious.”  Mr. Moye said that many of the prompts were entered the night before the expert’s deposition, and Mr. Moye uncovered a lot:

[The expert] also asked ChatGPT to “grade” his report (it got a 97/100), and “what are the 5 main things in my report the prosecution could attack and how do I defend them?” He then asked ChatGPT if his resume was sufficient to be an expert witness; “will prosecution go after me for never having been [an expert witness] before based on wording and how do I defend that?”

Additional details and part of the transcript are published in Mr. Koebler’s post. The post states: “In the examination at trial, Moye and Autenrieth agree that the submitted report is ‘90 to 85 percent ChatGPT.’” Id.

Mr. Moye said that “every lawyer needs to make sure their own experts aren’t generating work product in a way that’s insincere, and then knowing you can subpoena the prompts […]”  Id.

In response, the expert said that he has a body of work and more than 20 years of experience in the relevant field of gas detection.  The article states: “He did not explain why he used ChatGPT, but said ‘my opinions were put in there, and AI helped me to draft a straw man to build off of,’ and added ‘I put information and opinions in up front before it ever generated […] if the output wasn’t of my opinion or what I agreed to, I did alter it.’”

The post states: “The jury in the 3M case awarded more than $61 million to the plaintiffs, apportioning 30 percent of the responsibility to 3M and 70 percent to Watson Grinding.” Id.

Mr. Moye’s website states: “The Watson Grinding and Manufacturing explosion occurred when a propylene gas leak ignited at the facility in northwest Houston, sending a massive fireball through the neighborhood. The blast killed three employees, injured many others, and caused an estimated $200 million in property damage.”

Thanks to Jason Koebler for his post.  This one is based entirely on his post, which has much more detail.