Eric Martin | Does the Defend Trade Secrets Act Require Particularity in Pleading?
Background
A DNA sequencing company in California and a rotary die cutting manufacturer in North Carolina both sued former business partners for stealing trade secrets under the Defend Trade Secrets Act (DTSA). Both complaints described the alleged secrets in general terms: customer databases, proprietary processes, confidential business information. One case moved forward. The other was dismissed at the pleading stage, the threshold at which a court decides whether a complaint states a viable claim. The difference was not the lawyering, the facts, or the statutory text; it was the forum.
The DTSA, enacted in 2016 as an amendment to the Economic Espionage Act (EEA), Pub. L. No. 114-153, 18 U.S.C. § 1836, created the first private civil federal cause of action for trade secret misappropriation. The statute defines “trade secret” broadly under 18 U.S.C. § 1839(3), but says nothing about how specifically a plaintiff must identify those secrets in a complaint. That silence has produced a circuit split that was, in hindsight, predictable from the statute’s design.
The DTSA was drafted as a narrow graft onto the EEA, focused on creating a federal cause of action rather than building out the procedural infrastructure that state courts, particularly California, had spent decades developing to make trade secret litigation workable. The California Uniform Trade Secrets Act (CUTSA) requires a plaintiff to identify its trade secrets with “reasonable particularity” before commencing discovery, the pretrial phase in which the parties exchange evidence, Cal. Code Civ. Proc. § 2019.210. The DTSA imported none of it. Congress borrowed the what of trade secret protection (the substantive definition, misappropriation framework, and remedial scheme) without the how: the procedural tools that determine when and with what precision a plaintiff must say what was stolen.
Issue
Does the Defend Trade Secrets Act require a plaintiff to identify its alleged trade secrets with particularity at the pleading stage, or do general allegations suffice?
The Split
Fourth Circuit
The Fourth Circuit, in Sysco Machinery Corp. v. DCS USA Corp., 143 F.4th 222 (4th Cir. 2025), became the first federal appellate court to hold that the DTSA imposes a particularity requirement at the pleading stage. Sysco, a Taiwanese manufacturer of rotary die cutting machinery, sued its former North American distributor after three Sysco employees secretly formed a competing company and retained DCS to distribute rival machines. The complaint alleged that the defendants misappropriated Sysco’s “proprietary,” “confidential,” and “financial” information, but never identified which specific processes, designs, or data constituted the trade secrets at issue.
The Fourth Circuit affirmed dismissal, writing that Sysco had presented a “laundry list of vague company information” rather than identifying actual trade secrets. Sysco Machinery, 143 F.4th at 228. The court grounded its holding in the plausibility standard of Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009), and Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 570 (2007) (together, “Twiqbal”). It reasoned that, without knowing what the information actually is, a court cannot assess whether it qualifies as a trade secret.
The strongest version of the Fourth Circuit’s position frames this concern as a moral hazard: in technology disputes, where companies’ competitive advantages often overlap, a permissive standard lets plaintiffs file vague DTSA claims and use discovery to compel defendants to produce their own proprietary information, converting a trade secret suit into state-sponsored competitive intelligence. Given the breadth of § 1839(3), without a particularity screen, virtually any business information can be labeled a “trade secret” in a complaint. If a plaintiff genuinely cannot describe its trade secrets with specificity at the outset, it may not actually have trade secrets; it may have general business information it is retroactively characterizing as proprietary.
But the Fourth Circuit’s claim that it is merely applying Twiqbal does not survive closer inspection. Iqbal requires factual content making a claim plausible; it does not, in other contexts, require particularized identification of the claim’s subject matter at the complaint stage. The Fourth Circuit’s “sufficient particularity” standard, requiring description “at a level of detail that enables a defendant to delineate that which he is accused of misappropriating,” Sysco Machinery, 143 F.4th at 228, sounds less like Twiqbal’s general plausibility test and more like the heightened particularity requirement of Federal Rule of Civil Procedure 9(b) (which requires fraud claims to state the circumstances of the alleged fraud with specificity), imported into DTSA litigation without being named as such.
But the comparison is not merely illuminating. A court that requires DTSA plaintiffs to delineate their trade secrets before discovery has imposed something functionally equivalent to Rule 9(b): a heightened pleading requirement that neither the statute nor the Federal Rules authorize, and that in any other context would require rulemaking rather than judicial inference from statutory silence.
Months later, the Fourth Circuit applied its particularity requirement in Samuel Sherbrooke Corp., Ltd. v. Mayer, 159 F.4th 252, 256 (4th Cir. 2025), and found the plaintiff’s allegations sufficient. But in practice, meeting the standard requires a plaintiff to conduct a forensic investigation before filing, including imaging departing employees’ devices, reviewing system access logs, and cataloging the specific files, code repositories, or databases at issue. That front-loaded cost falls disproportionately on companies whose trade secrets are diffuse or collaborative.
Ninth Circuit
The Ninth Circuit reached the opposite conclusion in Quintara Biosciences, Inc. v. Ruifeng Biztech, Inc., 149 F.4th 1081 (9th Cir. 2025). Quintara, a California-based DNA sequencing company, sued a former business partner for misappropriating nine categories of trade secrets, including customer and vendor databases, proprietary software code, and marketing plans. The district court had ordered Quintara to identify its trade secrets with “reasonable particularity” before discovery, importing CUTSA’s standard into the federal proceeding, and then struck most of Quintara’s asserted secrets for insufficient identification. The Ninth Circuit reversed, holding that the DTSA contains no heightened identification requirement. Whether a plaintiff has adequately identified a trade secret is a question of fact for summary judgment or trial, not a threshold pleading requirement. Quintara Biosciences, 149 F.4th at 1089.
The court’s strongest move was textual, an expressio unius argument, the canon that the expression of one thing implies the exclusion of others: Congress knew how to impose an identification requirement (California’s legislature did so in § 2019.210) and chose not to include one in the federal statute. But this negative inference is weaker than it first appears. Congress modeled the DTSA on the Uniform Trade Secrets Act (UTSA), which also lacks CUTSA’s identification provision; § 2019.210 is a California-specific addition that most UTSA jurisdictions never adopted. The negative inference works only if Congress was aware of § 2019.210 and deliberately chose not to replicate it; if Congress was working from the UTSA template, the absence proves nothing about intent.
Nothing in the DTSA’s legislative history (three committee reports, floor statements from at least five members of Congress, and three associated hearings) addresses trade secret identification directly. The Ninth Circuit relied on the argument that requiring premature specificity forces a plaintiff to describe the very information it claims is secret. However, that tension is practically mitigated where protective orders, in camera review, and sealed filings ensure that identification occurs under standard protective mechanisms, not in public.
Still, the split is narrower than the opposing outcomes suggest, and its key dimension is one that neither circuit made explicit. What the Ninth Circuit actually held is more limited than it might appear. The court reversed a discovery sanction, not a motion to dismiss, and still requires “sufficient particularity” as a factual matter for proving trade secret ownership at summary judgment or trial. The real difference between the circuits is timing, not standard. Both require specificity; they disagree about when a plaintiff must provide it.
That distinction reframes the entire split: the question is not whether trade secrets must be identified with precision, but whether a complaint or discovery is the right vehicle for that identification. The Tenth Circuit had already applied the post-discovery approach in Double Eagle Alloys, Inc. v. Hooper, No. 24-5089 (10th Cir. Apr. 22, 2025), decided three months before Sysco Machinery. The facts were strikingly parallel (a former employee at a specialty metals distributor who downloaded 2,660 files before joining a direct competitor), but the Tenth Circuit evaluated identification at summary judgment, requiring the plaintiff to differentiate its alleged trade secrets from general industry knowledge only after discovery had run.
The timing question has concrete consequences. For instance, a company suing departing employees for misappropriation during an acquisition in the Ninth Circuit need not identify, at the outset, precisely which pre-acquisition secrets were taken, and can use discovery to refine its claims. In the Fourth Circuit, the same acquirer faces dismissal if its complaint does not specify the particular secrets at issue, a difficult task when another company’s engineers developed the IP.
The split also shapes how acquisition agreements are drafted. In a typical technology acquisition, the seller’s representations and warranties will include a schedule of material trade secrets as part of the IP representations under Article III (the section of a standard acquisition agreement containing the seller’s representations and warranties). Under the Fourth Circuit’s standard, the specificity required for that schedule may need to approach the specificity required to survive a motion to dismiss, giving buyers leverage to demand granular trade secret identification before signing rather than accepting a general representation that the target “owns valuable intellectual property.”
The identification question also reaches post-closing enforcement. Buyers in technology acquisitions typically require key employees to sign restrictive covenant agreements as a closing deliverable, and courts in many jurisdictions will not enforce those covenants unless the employer identifies a protectable business interest that the restriction is designed to protect. A buyer in the Fourth Circuit who cannot identify the target’s trade secrets with particularity at the complaint stage may also struggle to enforce the non-competes premised on those same secrets. In the Ninth Circuit, a broader representation may suffice for diligence, litigation, and enforcement alike.
Three circuits have now weighed in, and the Fourth Circuit stands alone in requiring identification at the pleading stage. The Ninth encompasses Silicon Valley and Seattle, the Fourth covers Northern Virginia and the Research Triangle, and the Tenth includes Denver’s energy and technology sectors. The divergence creates immediate incentives for forum shopping, the strategic selection of a favorable court in which to file. Consider this scenario: an engineer leaves a machine learning company in San Jose, joins a competitor in Reston, Virginia, and downloads proprietary training data before departing. The former employer can likely establish venue in either the Northern District of California or the Eastern District of Virginia under 28 U.S.C. § 1391(b)(2), since a substantial part of the events giving rise to the claim occurred in both districts. That choice of venue is now a choice of pleading standard. For cross-border acquirers whose trade secrets were developed by distributed engineering teams spanning multiple countries, the complexity compounds, since jurisdictions outside the United States impose their own identification standards at different procedural stages.
The gap is acute for AI companies, whose trade secrets resist the kind of discrete identification that the Fourth Circuit’s standard contemplates. An AI company’s trade secret may consist of the specific combination of a training dataset (including its composition and curation methodology), a model architecture, hyperparameter configurations derived from extensive experimentation, and the resulting trained weights. These elements are interdependent: describing any one in isolation may not capture the trade secret. Even in technical diligence, where a buyer’s engineers conduct code reviews, architecture walkthroughs, and data provenance assessments with the target’s full cooperation and access to source code repositories, identifying discrete protectable secrets within an AI system often takes weeks and yields inherently provisional conclusions. If a sophisticated buyer conducting privileged diligence cannot isolate what constitutes the trade secret, a plaintiff without access to the defendant’s systems cannot reasonably be expected to achieve that specificity at the complaint stage.
Looking Forward
No cert petition has been filed on the identification question in either Sysco Machinery or Quintara Biosciences. The Court has been selective with the DTSA, declining in February 2025 to review the Seventh Circuit’s extraterritoriality decision in Motorola Solutions, Inc. v. Hytera Commc’ns Corp., 108 F.4th 458 (7th Cir. 2024). But the emerging appellate consensus may matter more than a cert grant.
With the Tenth Circuit’s Double Eagle decision, the Ninth and Tenth Circuits both place identification post-discovery, and the Fourth Circuit stands alone in requiring it at the pleading stage. The First Circuit, covering Boston’s biotech and life sciences corridor, where pharmaceutical and research trade secrets pose similar identification challenges, has not yet addressed the question. If the Fourth Circuit’s approach remains isolated, the split may resolve through forum selection rather than Supreme Court intervention, as practitioners simply file elsewhere when venue permits.
The identification question is not the only DTSA issue generating circuit disagreement. The Fifth and Second Circuits have diverged on the availability of “avoided costs” as a measure of unjust enrichment damages (which strip a wrongdoer of gains obtained through misappropriation), with the Second Circuit rejecting such damages where trade secret value was not diminished, Syntel Sterling Best Shores Mauritius Ltd. v. TriZetto Grp., Inc., 68 F.4th 792, 811–12 (2d Cir. 2023), and the Fifth Circuit recognizing avoided costs as a valid theory before vacating a specific award as duplicative of a permanent injunction, Computer Sciences Corp. v. Tata Consultancy Services Ltd., 159 F.4th 429 (5th Cir. 2025). TCS filed a petition for certiorari in Computer Sciences in March 2026. The accumulation of unresolved federal questions under a statute barely a decade old confirms that DTSA jurisprudence is entering a period of active development.
If the Fourth Circuit’s particularity standard prevails nationally, it will reshape litigation from the outset. Plaintiffs will need forensic investigations before filing, discovery plans will narrow around specifically identified secrets, and companies whose intellectual property resists particularized description will face structural barriers to relief. If the post-discovery approach becomes the national rule, the DTSA will remain a plaintiff-friendly vehicle with deferred identification and broader discovery access.
In either scenario, the standard will ripple upstream into deal practice, shaping how companies catalog trade secrets, how acquirers conduct IP diligence, and how deal lawyers size indemnification baskets, the contractual limits on a seller’s post-closing liability for breaches of representations. The pattern is familiar. Congress creates a uniform federal remedy, courts disagree on the procedural mechanics the statute never specified, and the resulting circuit split produces exactly the jurisdictional fragmentation the statute was designed to eliminate. Until the Supreme Court weighs in, or more circuits do, the identification standard will remain a function of geography, and the DTSA’s promise of a uniform federal remedy will remain exactly what it was a decade ago: a promise.
Find Eric on LinkedIn!
