TAOS v. Renesas: Trade Secrets | Outside IP Counsel

The Federal Circuit dismantled a $48.8 million trade-secret disgorgement award on three fronts at once: who decides it, how to apportion among secrets, and how long the unjust-enrichment clock runs once reverse engineering becomes possible.

Texas Advanced Optoelectronic Solutions, Inc. v. Renesas Electronics America, Inc., 895 F.3d 1304 (Fed. Cir. 2018) (Appeal Nos. 2016-2121, -2208, -2235), is the rare decision that touches nearly every contested question in trade-secret damages in a single opinion. A jury in the Eastern District of Texas had awarded TAOS $48,783,007 in disgorgement of the defendant’s profits for trade-secret misappropriation, plus $10 million in exemplary damages, in a fight between two makers of ambient light sensors (the components that dim a phone screen in a dark room). The Federal Circuit vacated the monetary award in its entirety and remanded, holding that disgorgement was not for the jury to decide, that the award improperly lumped together secrets that did not all qualify, and that any recovery had to be confined to a head-start period bounded by lawful reverse engineering.

At a glance

Case: Texas Advanced Optoelectronic Solutions, Inc. v. Renesas Electronics America, Inc., 895 F.3d 1304, Appeal Nos. 2016-2121, -2208, -2235 (U.S. Court of Appeals for the Federal Circuit), on appeal from the Eastern District of Texas
Decision date: May 1, 2018. The panel then modified its opinion on panel rehearing and reissued it on July 9, 2018. The reissued opinion, reported at 895 F.3d 1304, superseded the original at 888 F.3d 1322 and is the operative one to cite
Jury award: $48,783,007 disgorgement of profits + $10 million exemplary damages for trade-secret misappropriation
Holdings: No Seventh Amendment jury-trial right on disgorgement (it is equitable); the unapportioned award could not stand where only one of three asserted secrets survived; disgorgement must be limited to a head-start period, which runs from the date the technology became properly accessible (here, by reverse engineering) for as long as it would have taken the defendant to recreate it in its own products
Outcome: Monetary award vacated and remanded for the district court to make equitable findings

Disgorgement is equitable, so the jury should not have set it

The first holding is procedural but consequential. TAOS sought disgorgement of the defendant’s profits, an unjust-enrichment remedy that asks not what the plaintiff lost but what the defendant gained. The Federal Circuit, applying the Seventh Amendment’s historical test, held there was no right to a jury determination of disgorgement for trade-secret misappropriation, because disgorgement of an infringer’s profits was historically an equitable remedy tried to the court, not to a law jury.

The practical upshot is that the $48.8 million figure the jury returned had no legal force as a damages verdict; on remand, the judge would have to make findings of fact and conclusions of law to fix any disgorgement amount. For litigants, the lesson is to plan the trial structure around the remedy sought: a plaintiff that wants the rhetorical and anchoring power of a jury number should pair disgorgement with a legal measure (actual loss or a reasonable royalty) that a jury can decide, rather than rely on equitable disgorgement alone.

Apportionment: you cannot recover for secrets you did not prove

TAOS had asserted three trade secrets, and the jury’s monetary award rested on all three without separating their respective contributions. On appeal, the structure narrowed dramatically: the court concluded that the protectable, misappropriated secret was the specific photodiode array structure using a 1:1 ratio of interleaved shielded and unshielded wells. Liability could not rest on either of the other two theories at all: the court held that the “Build vs. Buy” financial-information theory failed as a matter of law, and that the glass packaging roadmap theory lacked sufficient evidence.

Because the award was a single, unapportioned sum tied to all three theories, the court could not salvage it. TAOS’s damages expert had assigned all profits to the misappropriation of all three secrets without explaining which secret contributed what, so the court had no basis to conclude that the one surviving ground supported the entire award. It vacated the monetary award for trade-secret misappropriation and remanded for fresh factual findings. This is the apportionment principle in its starkest form: when damages are built on a bundle of asserted secrets and some fall away, an aggregate number that does not isolate the contribution of the surviving secret cannot stand. Plaintiffs who present secrets as an undifferentiated portfolio invite exactly this kind of total vacatur, a risk that has only grown as courts demand secret-by-secret proof.

The head-start clock and the limit of reverse-engineerable secrets

The most doctrinally rich holding concerns timing. The technology at issue could, in principle, be reverse engineered by proper means. Trade-secret law does not punish a competitor for eventually arriving at information it could have obtained legitimately; it remedies the head start the misappropriator gained by taking a shortcut. Accessibility by proper means, the court held, ended the information’s status as a protected secret, and secrecy protection terminated at the end of the period of time it would have taken the defendant, after that permissible discovery, to recreate the structure in its own products.

That splits the damages inquiry into two questions of fact, and the court remanded for findings on both: when did the secret become properly accessible to the defendant, and how long is the head-start period that follows? The accessibility date here did not turn on the defendant’s own laboratory work from scratch. TAOS had itself released a product embodying the design, so the design was open to reverse engineering by anyone; the court noted accessibility existed no later than January 2006, when the defendant actually did reverse-engineer the product, and perhaps as early as February 2005, when TAOS released it. Once the head-start period closes, the defendant’s continued profits are not attributable to the misappropriation, and unjust enrichment ends. The opinion makes the head-start theory not a soft equitable gloss but a hard ceiling on disgorgement, and it ties the ceiling to the defendant’s lawful alternatives, not to the plaintiff’s preferred recovery period.

Open questions

The 2018 opinion left the actual disgorgement figure unresolved, and with it the practical mechanics of measuring a head-start period for a technology that evolves quickly. Much of that has since been answered. On remand the district court found the secret properly accessible in January 2006, set a 26-month head start, and entered $8,546,000 in disgorgement plus $17,092,000 in exemplary damages. On the second appeal, AMS-OSRAM USA Inc. v. Renesas Electronics America, Inc., No. 22-2185 (Fed. Cir. Apr. 4, 2025), the panel reversed the accessibility date and held that a secret becomes properly accessible when it could have been reverse engineered, not when it actually was: February 28, 2005, once TAOS’s own product was on the market. It affirmed the 26-month duration, which put the end of the head start at April 28, 2007, and left the disgorgement award standing. So the timeline is fixed by both the plaintiff’s public release and the defendant’s own development record. What remains open is how granular that reconstruction must be. The 2018 opinion also sharpens, without fully resolving, the tension between the jury’s role and the court’s: when a case blends a legal measure (a royalty) with an equitable one (disgorgement), how should overlapping fact-finding be allocated so the two do not double-count or contradict each other?

Implications

Match the remedy to the fact-finder. Disgorgement is equitable and decided by the judge; if you want a jury number, plead and prove a legal measure (actual loss or reasonable royalty) alongside it.
Apportion from the outset. Tie damages proof to each asserted secret individually; an aggregate award resting on secrets that later fail is vulnerable to total vacatur.
The head start is the ceiling. Where information is reverse-engineerable, unjust-enrichment recovery ends once the secret was lawfully accessible and the defendant would have had time to recreate it; build (or attack) both dates as core damages facts.
Reverse-engineerability cuts both ways. It does not defeat liability, but it powerfully limits the recovery window. Defendants should develop it as a damages defense even after losing on liability.

Frequently asked questions

Why did the Federal Circuit say the jury could not award disgorgement? Disgorgement of a defendant’s profits is an equitable remedy that, historically, was tried to a court rather than a jury. Under the Seventh Amendment’s historical test, there was no jury-trial right on it, so the judge must determine the amount on remand.

What is a “head-start” damages period? It is the time advantage the misappropriator gained by stealing information instead of obtaining it lawfully, in this case by reverse engineering. Disgorgement is limited to that window, which runs from the date the secret became properly accessible by lawful means and lasts as long as it would have taken the defendant to recreate the technology in its own products. Profits earned after the window closes are not attributable to the misappropriation.

Why was the entire award vacated when one secret survived? Because the jury’s monetary award was a single, unapportioned sum based on all three asserted secrets. Liability could properly rest on only one of the three, and TAOS’s expert had assigned all profits to all three without separating them, so the court had no basis to conclude the surviving secret supported the entire award. It vacated and remanded for new findings.