Taranto, Circuit Judge.

Michael Oakes, Hunton Andrews Kurth LLP, Washington, DC, argued for plaintiff-appellant. Also represented by Ozzie Farres, Brian L. Saunders, Steven Leslie Wood ; David Parker, Richmond, VA.

Gabriel Bell, Latham & Watkins LLP, Washington, DC, argued for defendant-cross-appellant. Also represented by Tara Elliott, Rachel Weiner Cohen, Michael A. Morin.

Before Prost, Chief Judge, Reyna and Taranto, Circuit Judges.

Taranto, Circuit Judge. TecSec, Inc. brought this case in 2010 against several companies, including Adobe Inc., alleging that the companies directly and indirectly infringed claims of four TecSec patents. Aspects of the case have been before this court three times already. The present appeal involves Adobe only and several rulings of the district court, of which two are central. Specifically, before trial, in response to a motion in limine by Adobe, the court excluded all evidence of induced infringement from March 3, 2011, through the expiration of the patents at issue in October 2013. Earlier, the court had rejected Adobe's challenge to the asserted claims as ineligible under 35 U.S.C. § 101.

A jury found for TecSec on direct infringement, but not induced infringement; rejected Adobe's prior-art validity challenges; and awarded damages. The district court, though generally denying Adobe's post-trial motions, reduced the damages award to zero on the ground that there was no proof of any damages from direct infringement and the jury had rejected induced infringement.

TecSec appeals. It challenges the district court's motion-in-limine ruling, as well as certain jury instructions and the post-trial damages reduction. Adobe cross-appeals, challenging the district court's ruling on eligibility. We reverse the evidentiary ruling that eliminated TecSec's inducement case for a substantial period, and we reject Adobe's challenge to the district court's eligibility ruling. For those reasons, and others stated in this opinion, we reverse the judgment in part and remand for further proceedings on TecSec's claim of induced infringement.

I

A

TecSec owns U.S. Patent Nos. 5,369,702, 5,680,452, 5,717,755, and 5,898,781, the patents asserted in this case. The patents are entitled "Distributed Cryptographic Object Method" (the "DCOM patents") and claim particular systems and methods for multi-level security of various kinds of files being transmitted in a data network. See '702 patent, col. 3, lines 12–24; id. , col. 12, lines 2–16; id. , col. 12, line 45, through col. 13, line 20. In particular, the DCOM patents describe a method in which a digital object—e.g. , a document, video, or spreadsheet—is assigned a level of security that corresponds to a certain combination of access controls and encryption. Id. , col. 3, line 58, through col. 4, line 2; id. , col. 4, lines 18–25; id. , col. 5, lines 42–51. The encrypted object can then be embedded or "nested" within a "container object," which, if itself encrypted and access-controlled, provides a second layer of security. Id. , col. 4, lines 25–34.

Claims 1 and 8 of the '702 patent are representative of the asserted claims. Claim 1 recites:

1. A method for providing multi-level multimedia security in a data network, comprising the steps of:

A) accessing an object-oriented key manager;

B) selecting an object to encrypt;

C) selecting a label for the object;

D) selecting an encryption algorithm;

E) encrypting the object according to the encryption algorithm;

F) labelling the encrypted object;

G) reading the object label;

H) determining access authorization based on the object label; and

I) decrypting the object if access authorization is granted.

'702 patent, col. 12, lines 2–15. Whereas the subject of claim 1 is a method, the subject of claim 8 is a system that includes components—a "system memory," "an encryption algorithm module," "an object labelling subsystem," "a decryption algorithm module," and "an object label identification subsystem"—that carry out the steps of claim 1's method. Id. , col. 12, line 45, through col. 13, line 19.

B

In 2010, TecSec filed a complaint alleging infringement of the DCOM patents by Adobe, International Business Machines Corp. (IBM), and several other companies. On March 3, 2011, after staying the case as to all defendants except IBM, the district court construed several claim terms of the DCOM patents, including the term "multimedia." Based on those constructions, the district court granted IBM summary judgment of noninfringement. In January 2012, we summarily affirmed that noninfringement judgment without reaching certain issues, including the construction of "multimedia." TecSec, Inc. v. Int'l Business Machines Corp. , 466 F. App'x 882 (Fed. Cir. 2012).

The case then proceeded in district court against Adobe and other defendants. As relevant here, on April 23, 2012, TecSec and Adobe stipulated that, under the claim construction already adopted (which TecSec reserved the right to appeal), TecSec could not show that Adobe or the users of its accused Acrobat products infringed the DCOM patents, as alleged, through their actions involving portable document format documents (PDFs). The next day, the district court entered a judgment of noninfringement for Adobe. On October 2, 2013, we reversed the district court's construction of "multimedia" and remanded for further proceedings under the correct construction. TecSec, Inc. v. Int'l Business Machines Corp. , 731 F.3d 1336, 1345–49 (Fed. Cir. 2013) ( TecSec I ), cert. denied sub nom. Cisco Systems, Inc. v. TecSec, Inc. , 572 U.S. 1158, 134 S.Ct. 2698, 189 L.Ed.2d 756 (2014).

The DCOM patents expired on October 18, 2013, shortly after our TecSec I ruling. In October 2014, Adobe filed a motion for summary judgment of noninfringement, J.A. 165, and the district court granted the motion in May 2015, J.A. 27086–127. In August 2016, citing genuine disputes over material facts, we reversed that judgment and again remanded to the district court. TecSec, Inc. v. Adobe Systems Inc. , 658 F. App'x 570 (Fed. Cir. 2016) ( TecSec II ).

The case returned to the district court. In March 2017, Adobe moved for summary judgment of ineligibility under § 101. J.A. 1841–71. The district court, noting that the parties agreed that claims 1 and 8 of the '702 patent were representative for § 101 purposes, J.A. 4, denied the motion in May 2017. J.A. 3–13. In its ruling, the court stated that its rationale actually warranted "judgment in favor of [TecSec]" on Adobe's ineligibility challenge. J.A. 8.

In April 2018, after additional discovery, Adobe filed another motion for summary judgment of noninfringement, making at least one argument for lack of provable inducement of infringement. See ECF 1054 (motion); ECF 1123 at 10–11 (reply supporting motion). The district court denied the motion in June 2018. J.A. 10298–99. In August 2018, Adobe stipulated that "on at least one occasion," "there was direct infringement of the asserted claims" by an Adobe employee. J.A. 10307–08. It is common ground on appeal that this stipulation was referring to an occasion when an Adobe employee performed the steps of the method claims using an accused version of Adobe's Acrobat product, reflected in a February 2009 blog post on "packaging options for encrypted PDFs" that he wrote for an Adobe users' forum. See J.A. 24459–60.

In October 2018, as trial was approaching, Adobe filed an omnibus motion in limine, the first heading of which asked the court to "preclude argument, evidence, or testimony on Adobe's intent to induce or willfully infringe between March 3, 2011 and October 18, 2013, unless TecSec concedes admissibility of the March 3, 2011 Order (DX-28), Stipulation of Non-infringement (DX-29), and Judgment (DX-30)." J.A. 10429. Adobe argued that a "judgment or stipulation of non-infringement is highly relevant to an accused infringer's state of mind." Id. "If TecSec argues that Adobe had specific intent to induce infringement or willfully infringe during this period," Adobe stated, "TecSec opens the door to the admissibility of this exculpatory evidence"—which, Adobe asserted (without citing evidence), it reasonably relied on, so that "there can be no specific intent to induce infringement during [the] roughly 31-month period" from March 2011 to October 2013. Id. TecSec argued that intent was a fact question for the jury, and it disagreed with Adobe's submission that allowing proof of inducement during the period at issue required admission of the three documents. J.A. 10849. TecSec did not indicate that it would decline to proceed with its inducement case for the period at issue if that evidence was to be admitted.

At an oral argument on the omnibus motion on Friday, November 9, 2018, the district court indicated that it intended to preclude "any reference" to Adobe's post-March 2011 conduct, making admission of the three documents immaterial. ECF 1294, p. 35; id. at pp. 3–4, 34–37. The court stated its intent in that respect equally for the inducement contention and the willfulness contention. Id. On November 13, 2018, after the Veterans Day weekend but before any ruling by the court, TecSec filed a supplemental brief stating expressly that even if the district court allowed Adobe to present evidence of Adobe's lack of intent, "the Court should not preclude TecSec from presenting evidence on Adobe's intent to induce infringement." TecSec, Inc.'s Suppl. Br. Regarding Def. Adobe Inc.'s Mot. in Limine #1 at 7, TecSec, Inc. v. Adobe, Inc. , No. 1:10-cv-00115-LO-TCB (E.D. Va. Nov. 13, 2018), ECF 1283-1 (TecSec's Supp. MIL Br.). The district court, TecSec argued, could not "resolve a substantive factual dispute through a limine ruling before the evidence is presented at trial." Id.

On November 21, 2018, the district court granted Adobe's motion and excluded all evidence of induced infringement after March 3, 2011. J.A. 26–29. The court based its decision on what Adobe treats as two independent grounds. First, the court held that, after the favorable 2011 claim construction, which was reasonable even though it was later reversed, it was impossible as a matter of law for Adobe to have had the requisite intent to induce third-party infringement. J.A. 27. Second, the court stated that, while allowing TecSec to argue inducement of infringement without permitting Adobe to introduce the three documents at issue would highly prejudice Adobe, admitting that evidence would be prejudicial to TecSec and confuse the jury. J.A. 28–29. The court therefore barred the presentation of all inducement evidence concerning the period after March 3, 2011. Id.

The case proceeded to a jury trial, which began on December 10, 2018. On December 12, 2018, TecSec stipulated that, because it had failed to mark its product in accordance with 35 U.S.C. § 287, it would not seek damages for any infringement that took place before the filing of its complaint. J.A. 11998. Accordingly, the trial was limited to TecSec's ownership of the patents; direct infringement by Adobe from February 2010 to October 2013; induced infringement by Adobe from February 2010 to March 3, 2011; Adobe's anticipation and obviousness challenges; and damages. See J.A. 38–44.

At trial, TecSec offered the deposition testimony of George Johnson, an independent technology consultant, as evidence of direct infringement by third parties allegedly induced by Adobe. Mr. Johnson had stated that, in April 2011, he provided Acrobat users with guidance on how to use certain features. The court excluded certain parts of Mr. Johnson's testimony, stating that the period for inducement ended on March 3, 2011, and that TecSec could not submit evidence of later conduct. J.A. 12429–30.

At the conclusion of trial, the court gave the jury two instructions relevant to this appeal. Jury Instruction No. 25 highlighted Adobe's stipulation that the 2009 employee blog post constituted direct infringement, and it noted that this act had occurred before TecSec filed its complaint in 2010. J.A. 12076; see J.A. 10307–08. Jury Instruction No. 28 stated that "mere inaction, or a failure to stop or prevent infringement, does not constitute inducement." J.A. 12079.

On December 19, 2018, the jury, besides confirming TecSec's ownership of the patents, J.A. 39, found that Adobe had directly infringed all asserted claims, but that Adobe had not committed inducement of infringement, J.A. 40–41. The jury also found that Adobe had failed to establish that the asserted claims were invalid under the prior art. J.A. 42–43. The jury awarded TecSec $1.75 million in damages. J.A. 44.

The district court then ruled on Adobe's post-trial motion. It declined to disturb the verdicts on direct infringement and invalidity, J.A. 53–55, but it vacated the jury's damages award and set damages to zero, J.A. 55–60. The court explained that TecSec's only evidence of damages related to sales of Acrobat software to customers and that such evidence could support damages only for inducing infringement—which the jury found had not occurred. Id.

The district court thereafter entered a final judgment as to Adobe under Fed. R. Civ. P. 54(b). J.A. 2. TecSec timely appealed, and Adobe timely cross-appealed. We have jurisdiction under 28 U.S.C. § 1295(a)(1).

II

In general, we apply regional-circuit law to evidentiary determinations and jury instructions. Barry v. Medtronic, Inc. , 914 F.3d 1310, 1320 (Fed. Cir. 2019). The Fourth Circuit reviews both types of rulings for an abuse of discretion. See Campbell v. Boston Sci. Corp. , 882 F.3d 70, 77 (4th Cir. 2018) (evidentiary determinations); Chaudhry v. Gallerizzo , 174 F.3d 394, 408 (4th Cir. 1999) (jury instructions). But "where a district court rules, as a matter of patent law, that a party is precluded from introducing evidence, we apply Federal Circuit law and review the district court's ruling de novo." Sulzer Textil A.G. v. Picanol N.V. , 358 F.3d 1356, 1363 (Fed. Cir. 2004). We review de novo a jury instruction that concerns patent law. Bettcher Indus., Inc. v. Bunzl USA, Inc. , 661 F.3d 629, 638 (Fed. Cir. 2011). We review a judgment as a matter of law de novo. Barry , 914 F.3d at 1320 ; Huskey v. Ethicon, Inc. , 848 F.3d 151, 156 (4th Cir. 2017). Whether a patent claim is eligible for patenting under 35 U.S.C. § 101 is a question of law, to which underlying facts may be relevant. Berkheimer v. HP Inc. , 881 F.3d 1360, 1365 (Fed. Cir. 2018).

A

We begin with TecSec's appeals of the district court's motion-in-limine ruling and the court's related exclusion of Mr. Johnson's deposition testimony. We reverse the district court's decision to exclude evidence of inducement of infringement after March 3, 2011, and the related exclusion of Mr. Johnson's testimony. We remand for further proceedings on TecSec's inducement claim as to the period after March 3, 2011.

The district court set forth two reasons for ruling that there could be no proof of post-March 3, 2011 inducement of infringement. First, the district court explained that the March 3, 2011 claim construction and stipulation of noninfringement made it legally impossible for Adobe to have had, after that date, the knowledge—of the infringing character of the allegedly induced customer actions—required for inducement. J.A. 27–28. Because such knowledge was legally precluded, the district court reasoned, Adobe could not be liable for induced infringement. Id . Second, the district court stated that allowing either party to present evidence would substantially prejudice the other party and would confuse the jury. J.A. 28–29. We reject both rationales for the motion-in-limine ruling.

1

A defendant is liable for "induced infringement under § 271(b)" if the defendant took certain affirmative acts to bring about the commission by others of acts of infringement and had "knowledge that the induced acts constitute patent infringement." Global-Tech Appliances, Inc. v. SEB S.A. , 563 U.S. 754, 765–66, 131 S.Ct. 2060, 179 L.Ed.2d 1167 (2011) ; see Commil USA, LLC v. Cisco Systems, Inc. , 575 U.S. 632, 135 S. Ct. 1920, 1928, 191 L.Ed.2d 883 (2015) ; Power Integrations, Inc. v. Fairchild Semiconductor Int'l, Inc. , 843 F.3d 1315, 1331 (Fed. Cir. 2016). The intent element requires "knowledge that the induced acts constitute patent infringement," which can be established by a proper finding of "willful blindness." Global-Tech , 563 U.S. at 766–71, 131 S.Ct. 2060 ; see Commil , 135 S. Ct. at 1926–28 (reiterating requirement of knowledge of infringing character of induced conduct, not just knowledge of patent). This intent element was the basis for the district court's primary rationale for its motion-in-limine ruling, a ruling that assumed that the induced conduct was in fact infringing.

The intent standard focuses on, and can be met by proof of, the defendant's subjective state of mind, whether actual knowledge or the subjective beliefs (coupled with action to avoid learning more) that characterizes willful blindness. Global-Tech , 563 U.S. at 769, 131 S.Ct. 2060. As a logical matter, a defendant may have the liability-supporting subjective state of mind even if a person could believe, with objective reasonableness (though wrongly), that the induced conduct was not infringing. To make the point in terms of this case, Adobe may have had the requisite knowledge of infringement if it believed (as we ultimately held in 2013) that the March 3, 2011 claim construction was incorrect, even if that construction was objectively reasonable.

We have so ruled as to willfulness—an issue that Adobe, in its motion, and the district court, in its ruling, treated together with inducement. Specifically, we have held that, under Halo Electronics, Inc. v. Pulse Electronics, Inc. , ––– U.S. ––––, 136 S. Ct. 1923, 1933, 195 L.Ed.2d 278 (2016), a finding of willfulness may rest on the subjective bad faith of the infringer even if it would be objectively reasonable to view the conduct at issue as non-infringing. WesternGeco LLC v. ION Geophysical Corp. , 837 F.3d 1358, 1362 (Fed. Cir. 2016), reinstated 913 F.3d 1067, 1075 (Fed. Cir. 2019). That logic applies equally to the intent element of inducement, as we concluded in our non-precedential decision in Smith & Nephew Inc. v. Arthrex, Inc. , 603 F. App'x 981 (Fed Cir. 2015), where we explained that a district court's prior judgment of noninfringement, and statement to the parties that the plaintiff's inducement claim was objectively weak, did not prevent the plaintiff, as a matter of law, from proving the subjective intent element of induced infringement, which is a fact issue. Id. at 989–90 ; see also Sanofi v. Watson Labs. Inc. , 875 F.3d 636, 645 (Fed. Cir. 2017) (factual issue); AstraZeneca LP v. Apotex, Inc. , 633 F.3d 1042, 1056, 1059–60 (Fed. Cir. 2010) (same).

The Supreme Court's conclusion about subjective beliefs and legal reasonableness in another context supports this conclusion. In Cheek v. United States , 498 U.S. 192, 111 S.Ct. 604, 112 L.Ed.2d 617 (1991), the Court reviewed the conviction of a taxpayer charged with willful failure to pay taxes, which requires "knowing" that the law required the payment not made. Id. at 194 n.2, 199–201, 111 S.Ct. 604. The Court held that, even though it would be not just incorrect but unreasonable as a matter of law for the taxpayer to believe that the statute did not require payment, the jury had to be permitted to decide whether the taxpayer in good faith believed the incorrect, legally unreasonable view that no tax was due. Id. at 202–04, 111 S.Ct. 604. Thus, the legally material question of what the defendant subjectively believed was not answered as a matter of law even by the objective unreasonableness of that belief. Likewise, the question whether the defendant subjectively had what turned out to be the correct belief (here, that Adobe believed that the claims meant what we eventually held in 2013) is not answered as a matter of law by the objective reasonableness of a contrary belief.

The district court in this case erred when it concluded as a matter of law that, after the district court's (later reversed) claim construction on March 3, 2011, Adobe "lacked the requisite intent to induce infringement." J.A. 27. The court explained that it had "reasonably, though erroneously, ruled in Adobe's favor on infringement," and, accordingly, Adobe was entitled to rely on that ruling as proof that the relevant induced acts were non-infringing. Id. That reasoning makes dispositive what Adobe, with objective reasonableness, could have believed. The Global-Tech inducement standard, however, can be met by proof of what Adobe in fact subjectively believed. For example, Adobe might have believed that the March 2011 claim-construction ruling was erroneous (though reasonable) and would likely be reversed (as it was in 2013). Thus, the district court's March 2011 ruling "could, at most, create a factual question, not an entitlement to a no-knowledge finding as a matter of law." Smith & Nephew , 603 F. App'x at 989–90. For those reasons, we conclude that the district court legally erred in its primary rationale for ruling out inducement after March 3, 2011, namely, that the claim-construction ruling of that date furnished an objectively reasonable basis for a belief that use of the accused products did not infringe, even if Adobe did not have such a belief.

2

The district court's second ground for its motion-in-limine ruling was that "allowing TecSec to argue [post-March 3, 2011] induced or willful infringement ... would taint the trial and any verdict with undue prejudice and juror confusion." J.A. 28. Whether we review that ruling de novo or for an abuse of discretion, we conclude that the district court erred in relying on this rationale to preclude TecSec from proving inducement after March 3, 2011.

In defending the district court's rationale, Adobe invokes Rule 403 of the Federal Rules of Evidence, which provides that a court "may exclude relevant evidence if its probative value is substantially outweighed by a danger of," among other factors, "unfair prejudice" or "confusing the issues." But the district court in this case went beyond excluding a single piece, or even a fixed set, of evidence and leaving TecSec to present such other relevant, admissible evidence as it may have on inducement in the period at issue. Instead, the court foreclosed the introduction of any evidence on the issue of post-March 3, 2011 inducement of infringement. This is materially different from any application of Rule 403 Adobe has identified. Cf. General Dynamics Corp. v. United States , 563 U.S. 478, 484–85, 131 S.Ct. 1900, 179 L.Ed.2d 957 (2011) (distinguishing case involving "purely evidentiary dispute" over state-secret privilege, where "the privileged information is excluded and the trial goes on without it," from case involving foreclosure of a claim).

The district court reasoned that if TecSec was allowed to argue inducement of infringement after that date, Adobe would be prejudiced were it not allowed to introduce the March 3, 2011 claim construction and the stipulation by TecSec that, under that construction, Adobe's customers did not directly infringe. J.A. 28. The court then concluded that "it would also be substantially and unduly prejudicial and confusing for the jury to see a prior ruling in this case and TecSec's stipulation that Adobe's products did not infringe." J.A. 28–29. On that basis, the district court precluded TecSec from offering any proof of post-March 3, 2011 inducement.

Adobe has not cited any authority that supports such foreclosure of liability for a given period in the circumstances present here. Any conclusion that post-March 3, 2011 inducement could not be fairly tried would have required consideration of what other admissible evidence TecSec had that was relevant to the intent element of inducement, as well as of the possible measures for effectively but fairly reducing jury confusion. But the district court did not undertake, and because of the limited scope of Adobe's motion in limine was in no position to undertake, that required consideration.

Adobe's motion, besides what it suggested on the legal-impossibility point, was limited to arguing for Adobe's ability to introduce the claim-construction ruling and TecSec's stipulation. It did not call for TecSec to identify all evidence of intent—as a motion for summary judgment would have done—so that the district court could determine if there was a triable issue of post-March 3, 2011 inducement of infringement. See J.A. 10429. TecSec therefore properly limited its response to arguing against the admission of the construction and stipulation. See J.A. 10849. On the motion as framed, the district court lacked the basis to consider the totality of TecSec's post-March 3, 2011 inducement evidence to determine whether that issue could fairly be tried. And the court did not consider the totality of such evidence. Nor, in taking the extraordinary action of foreclosing liability for a substantial period, did the district court set forth an analysis of potential means, such as cautionary jury instructions, that could reduce jury confusion and allow a fair trial considering the totality of relevant evidence.

Adobe cites several cases to support the proposition that "there are circumstances where, for willfulness or inequitable conduct, there can be no intent as a matter of law." Appellant's Br. at 28. None of those cases support the ruling in this case. The cited cases all involved ordinary motions for summary judgment that, with proper procedural protections, duly enabled the court to consider all evidence relevant to the particular liability issue. See Ferring B.V. v. Barr Labs., Inc. , 437 F.3d 1181, 1190–94 (Fed. Cir. 2006) (finding lack of intent for inequitable conduct after consideration of evidence on summary judgment); Intercontinental Great Brands LLC v. Kellogg N. America Co. , 869 F.3d 1336, 1352 (Fed. Cir. 2017) (same). By contrast, Adobe's motion in limine did not call for, or provide a basis for, consideration of all relevant material evidence. Without such consideration, the district court could not properly resolve the issue as a matter of law.

As Adobe acknowledges in this court, see Cross-Appellant's Principal and Response Br. at 15, 20, when the district court stated that it would be "substantially and unduly prejudicial and confusing for the jury to see" the claim-construction ruling and TecSec's stipulation, J.A. 28–29, the court was determining that the evidence Adobe was entitled to introduce would be "unduly prejudicial" to TecSec . But TecSec never stated that, if that evidence had to be admitted at Adobe's behest, TecSec would drop its inducement claim for the post-March 3, 2011 period. To the contrary, after the district court opined at the oral argument that admitting that evidence would be "far too prejudicial," ECF 1294 at 36, TecSec reiterated its intent to present the inducement case even if that evidence were admitted. TecSec's Supp. MIL Br. (ECF 1283-1) at 7. The court furnished no justification for making that determination itself instead of leaving to TecSec the determination that the prejudice to TecSec was too severe for the issue to be tried at all.

Accordingly, we reverse the district court's decision to preclude TecSec from introducing evidence of post-March 3, 2011 inducement of infringement.

3

TecSec's appeal of the exclusion of certain aspects of Mr. Johnson's deposition testimony is closely tied to the motion-in-limine ruling. In the testimony at issue, Mr. Johnson stated that, in April 2011, he had posted responses on an Adobe forum explaining how to use Acrobat products in ways that TecSec alleged were infringing. At trial, Adobe objected to the introduction of that testimony, arguing that the testimony "runs afoul of the Court's [motion-in-limine] ruling on eliciting alleged infringement [post-March 3, 2011]." J.A. 12427. The district court agreed with Adobe, excluding the testimony because "it's outside of the timeframe and it is clearly inducement evidence." J.A. 12429–30. Because we conclude that the district court erred in its motion-in-limine ruling that was the basis for the exclusion of the Johnson testimony at issue, we reverse that evidentiary exclusion as well.

B

TecSec challenges Jury Instruction Nos. 25 and 28. We find no reversible error in either instruction, asking whether TecSec has "establish[ed] that the instructions were legally erroneous and that the errors had a prejudicial effect." Bettcher , 661 F.3d at 638.

1

In Jury Instruction No. 25, the district court told the jury that Adobe had admitted that a certain sequence of steps ("create" a file, "attach" it to a second file, "encrypt" the second file, "decrypt" the first file) constituted direct infringement, that Adobe had admitted that "this occurred on at least one occasion, by Adobe itself," and that the "parties agree that the one time for which Adobe has admitted infringement—that is Adobe's use in connection with creating a February 2009 blog post—occurred before TecSec filed its lawsuit in February 2010." J.A. 12076 (emphases added). That instruction mirrors Adobe's pre-trial stipulation, differing only in its specific reference to the February 2009 blog post. See J.A. 10307.

TecSec challenges the district court's inclusion of the reference to the February 2009 blog post, stating that "TecSec never agreed that Adobe only admitted to a single act of infringement" and that the "instruction is misleading because it implies that, in the context of the district court's narrow damages window, Adobe's 2009 blog is irrelevant." Appellant's Br. at 43–44. We reject the challenge.

TecSec relies on a single piece of evidence as showing a broader Adobe admission, namely, Adobe's stipulation that it had "installed for its own internal use at least one copy of each version of the accused products on a computer having system memory." J.A. 11994 (emphasis added). But Jury Instruction No. 25 is best understood to tie "the one time" language, to which TecSec objects, to what the immediately preceding two sentences discuss, which is on its face performance of a method . Installation, by contrast, could constitute infringement only of the system claims . See J.A. 12769 (Trial Tr. at 534:4–5). The installation action thus is simply outside the scope of, and does not contradict, Jury Instruction No. 25—which left TecSec free to argue that there were instances of Adobe-admitted infringement of the system claims. Moreover, the reference to the timing of the blog post is accurate and does not itself imply anything about the legal relevance of the timing. We therefore find no reversible error in Jury Instruction No. 25, without foreclosing the district court, in any new trial on inducement after March 3, 2011, from considering whether the instruction should be modified.

2

In Jury Instruction No. 28, the district court told the jury that inducement of infringement requires "an affirmative act to encourage infringement" and that "[e]vidence of mere inaction, or a failure to stop or prevent infringement, does not constitute inducement." J.A. 12079. That language mirrors our statement in Tegal Corp. v. Tokyo Electron Co . that "[a]ctively inducing ... requires an affirmative act of some kind." 248 F.3d 1376, 1378 (Fed. Cir. 2001) (internal quotation marks omitted). TecSec challenges the instruction on the ground that it opened the door for Adobe to imply that TecSec had accused Adobe of inaction only. Specifically, TecSec argues, Adobe "exploited the improper instruction during closing" arguments when Adobe characterized its failure to remove the 2009 blog post as mere inaction. Appellant's Br. at 45; see J.A. 13544. We reject this challenge.

TecSec does not contest the accuracy of Jury Instruction No. 28 as a statement of the law, and it has not shown that instruction was misleading. TecSec believes that Adobe's conduct to have been more than mere inaction, but the instruction left TecSec free to so argue. As for Adobe's closing argument, TecSec forfeited a challenge to the argument by failing to object contemporaneously. See United States v. Cone , 714 F.3d 197, 217 (4th Cir. 2013).

C

TecSec challenges the district court's reduction of the jury's damages award to zero. We reject this challenge.

TecSec argues that because the jury found direct infringement of each asserted claim, 35 U.S.C. § 284 entitles TecSec to a non-zero reasonable royalty. Specifically, TecSec highlights § 284's statements that "[u]pon finding for the claimant the court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty" and that "[w]hen the damages are not found by a jury, the court shall assess them." 35 U.S.C. § 284. Taken together, TecSec contends, those statements require an award of damages greater than zero in all cases where the jury finds infringement.

The statute does not require an award of damages if none are proven that adequately tie a dollar amount to the infringing acts. We have explained that "a patent owner may waive its right to a damages award when it deliberately abandons valid theories of recovery in a singular pursuit of an ultimately invalid damages theory." Promega Corp. v. Life Techs. Corp. , 875 F.3d 651, 666 (Fed. Cir. 2017). More generally, we have observed that there can be an award of no damages where "none were proven." Gustafson, Inc. v. Intersystems Indus. Prods., Inc. , 897 F.2d 508, 509–10 (Fed. Cir. 1990) ; cf. Gadsden Indus. Park, LLC v. United States , 956 F.3d 1362, 1372 (Fed. Cir. 2020) (Takings Clause does not require an award of compensation without adequate proof). Although we have not upheld a zero royalty rate in a case with an affirmative infringement finding—and have stated that it is "unlikely" that a hypothetical negotiation would result in a zero royalty rate—we have previously stated that "in a case completely lacking any evidence on which to base a damages award, the record may well support a zero royalty award." Apple, Inc v. Motorola, Inc. , 757 F.3d 1286, 1328 n.7 (Fed. Cir. 2014).

In this case, TecSec presented no evidence of damages caused by Adobe's direct infringement, which was the only form of infringement that the jury found Adobe to have committed. J.A. 58. TecSec did not, for example, provide any evidence regarding the number of Adobe employees in the United States or the number of times that Adobe employees installed Acrobat. Id. TecSec's only damages evidence relied on "sales of accused products." E.g. , J.A. 12840, 12842–44, 12854–56, 12866; see also J. A. 13499–500 (TecSec's closing argument, stating, "all that TecSec is asking for here, a reasonable royalty based on [Adobe's] sales of Acrobat products"). But the district court found, and TecSec does not dispute, that direct infringement occurs only after Acrobat is installed. J.A. 57. Sales of Acrobat, therefore, cannot be a measure of damages for direct infringement by Adobe—sales could serve here as a measure of damages only for TecSec's indirect infringement theory, which relies on infringement by Adobe customers. But the jury found no indirect infringement.

On appeal, TecSec tries to rationalize the jury's award by theorizing ways the jury could have used sales data to derive a value for Adobe's conduct. For instance, TecSec argues that the jury could have decided that sales were a good proxy for the value of TecSec's patented invention to Adobe's development of Acrobat. Appellant's Opening Br. at 59–60. But this theory, like the others that TecSec offers, is ultimately speculative and insufficiently grounded in evidence. The district court therefore committed no error in concluding that the jury's damages award was not supported by the evidence. J.A. 57–58.

III

The district court denied Adobe's motion for summary judgment that the asserted claims are ineligible under 35 U.S.C. § 101, and in its opinion, the court stated that its conclusion required judgment for TecSec rejecting the eligibility challenge. Adobe now cross-appeals that ruling. It does not object to the district court's turning the denial of summary judgment for Adobe into summary judgment for TecSec. Adobe thus accepts that the dispositive ruling against its eligibility challenge should be judged on the record made on its motion for summary judgment. We affirm the district court's § 101 ruling.

A

Section 101 provides that "[w]hoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title." 35 U.S.C. § 101. But this provision "contains an important implicit exception: Laws of nature, natural phenomena, and abstract ideas are not patentable." Alice Corp. Pty. Ltd. v. CLS Bank Int'l , 573 U.S. 208, 216, 134 S.Ct. 2347, 189 L.Ed.2d 296 (2014). In Alice , the Supreme Court explained that a "claim falls outside § 101 where (1) it is ‘directed to’ a patent-ineligible concept, i.e. , a law of nature, natural phenomenon, or abstract idea, and (2), if so, the particular elements of the claim, considered ‘both individually and "as an ordered combination," ’ do not add enough to ‘ "transform the nature of the claim" into a patent-eligible application.’ " SAP America, Inc. v. InvestPic, LLC , 898 F.3d 1161, 1166–67 (Fed. Cir. 2018) (citing Alice , 573 U.S. at 217, 134 S.Ct. 2347 ).

We have approached the Step 1 "directed to" inquiry by asking "what the patent asserts to be the ‘focus of the claimed advance over the prior art.’ " Solutran, Inc. v. Elavon, Inc. , 931 F.3d 1161, 1168 (Fed. Cir. 2019) (quoting Affinity Labs of Texas, LLC v. DIRECTV, LLC , 838 F.3d 1253, 1257 (Fed. Cir. 2016) ); see Trading Techs. Int'l Inc. v. IBG LLC , 921 F.3d 1378, 1384 (Fed. Cir. 2019). In conducting that inquiry, we "must focus on the language of the Asserted Claims themselves," Synopsys, Inc. v. Mentor Graphics Corp. , 839 F.3d 1138, 1149 (Fed. Cir. 2016), "considered in light of the specification," Enfish, LLC v. Microsoft Corp. , 822 F.3d 1327, 1335 (Fed. Cir. 2016). See also Chamberlain Group, Inc. v. Techtronic Indus. Co. , 935 F.3d 1341, 1346 (Fed. Cir. 2019) ; ChargePoint, Inc. v. SemaConnect, Inc. , 920 F.3d 759, 766 (Fed. Cir. 2019) ; In re TLI Communications LLC Patent Litigation , 823 F.3d 607, 611–12 (Fed. Cir. 2016) ; Ariosa Diagnostics, Inc. v. Sequenom, Inc. , 788 F.3d 1371, 1376 (Fed. Cir. 2015). And we have reiterated the Supreme Court's caution against "overgeneralizing claims" in the § 101 analysis, explaining that characterizing the claims at "a high level of abstraction" that is "untethered from the language of the claims all but ensures that the exceptions to § 101 swallow the rule." Enfish , 822 F.3d at 1337 ; see Solutran , 931 F.3d at 1167–68 ; McRO, Inc. v. Bandai Namco Games America, Inc. , 837 F.3d 1299, 1313 (Fed. Cir. 2016) (explaining that courts " ‘must be careful to avoid oversimplifying the claims’ by looking at them generally and failing to account for the specific requirements of the claims").

"In cases involving software innovations, this inquiry often turns on whether the claims focus on specific asserted improvements in computer capabilities or instead on a process or system that qualifies an abstract idea for which computers are invoked merely as a tool." Uniloc USA, Inc. v. LG Electronics USA, Inc. , 957 F.3d 1303, 1306–07 (Fed. Cir. 2020) (citing Customedia Techs., LLC v. Dish Network Corp. , 951 F.3d 1359, 1364 (Fed. Cir. 2020), and Finjan, Inc. v. Blue Coat Systems, Inc. , 879 F.3d 1299, 1303 (Fed. Cir. 2018) ); see also BSG Tech LLC v. Buyseasons, Inc. , 899 F.3d 1281, 1285–86 (Fed. Cir. 2018) ; Core Wireless Licensing S.A.R.L. v. LG Electronics, Inc. , 880 F.3d 1356, 1361–62 (Fed. Cir. 2018). "[S]oftware can make patent-eligible improvements to computer technology, and related claims are eligible as long as they are directed to non-abstract improvements to the functionality of a computer or network platform itself." Uniloc , 957 F.3d at 1309 ; see Customedia Techs. , 951 F.3d at 1364 (collecting cases); Enfish , 822 F.3d at 1335. We have found claims directed to such eligible matter in a number of cases where we have made two inquiries of significance here: whether the focus of the claimed advance is on a solution to "a problem specifically arising in the realm of computer networks" or computers, DDR Holdings, LLC v. Hotels.com, L.P. , 773 F.3d 1245, 1257–58 (Fed. Cir. 2014) ; and whether the claim is properly characterized as identifying a "specific" improvement in computer capabilities or network functionality, rather than only claiming a desirable result or function, Uniloc , 957 F.3d at 1306, 1308–09 ; see Enfish , 822 F.3d at 1135–36 ; see also , e.g. , Packet Intelligence LLC v. NetScout Systems, Inc. , 965 F.3d 1299, 1309–10 (Fed. Cir. 2020) ; Ericsson Inc. v. TCL Communication Tech. Holdings Ltd. , 955 F.3d 1317, 1328 (Fed. Cir. 2020) ; Koninklijke KPN N.V. v. Gemalto M2M GmbH , 942 F.3d 1143, 1150 (Fed. Cir. 2019) ; SRI Int'l, Inc. v. Cisco Systems, Inc. , 930 F.3d 1295, 1303–04 (Fed. Cir. 2019) ; Ancora Techs., Inc. v. HTC America, Inc. , 908 F.3d 1343, 1347–49 (Fed. Cir. 2018) ; Visual Memory LLC v. NVIDIA Corp. , 867 F.3d 1253, 1258–59 (Fed. Cir. 2017) ; Intellectual Ventures I LLC v. Capital One Financial Corp. , 850 F.3d 1332, 1342 (Fed. Cir. 2017) ; McRO , 837 F.3d at 1313–16.

For example, in SRI , we found to be eligible claims directed to "a specific technique—using a plurality of network monitors that each analyze specific types of data on the network and integrating reports from the monitors—to solve a technological problem arising in computer networks: identifying hackers or potential intruders into the network." 930 F.3d at 1303. The claims focused on a solution to a computer-specific problem—uniquely difficult-to-track, large-scale attacks caused by the decentralized nature of computer networks—and a concrete improvement to network functionality—the deployment of specific monitors to collect specific types of data. Id. In Ancora , we upheld claims requiring use of a modifiable portion of particular computer capacity to prevent running software on an unauthorized computer. 908 F.3d at 1345. We noted that "[i]mproving security ... can be a non-abstract computer-functionality improvement if done by a specific technique that departs from earlier approaches to solve a specific computer problem"—there, the particular difficulty of controlling unauthorized use of software. Id. at 1348. In Data Engine Technologies LLC v. Google LLC , we held that claims reciting "a specific method for navigating through three-dimensional electronic spreadsheets" were patent eligible because the claimed advance "improv[ed] computers' functionality as a tool able to instantly access all parts of complex three-dimensional electronic spreadsheets"—a form of information-organization unique to computers. 906 F.3d 999, 1007–08 (Fed. Cir. 2018). And in Finjan , we held to be eligible claims directed to a specific "behavior-based virus scan" that provided a "flexible and nuanced virus filtering" to help solve the problem of security on networked computers. 879 F.3d at 1304–05.

Other illustrative cases lie on the ineligibility side of the line. In Ericsson , we held to be ineligible claims describing "a system and method for controlling access to a platform for a mobile terminal for a wireless telecommunications system." 955 F.3d at 1325–26. The claims, we explained, "merely make generic functional recitations that requests are made and then granted." Id. at 1328. In Intellectual Ventures I LLC v. Symantec Corp. , we held ineligible claims directed to using computers to screen telephone calls for viruses. 838 F.3d 1307, 1319 (Fed. Cir. 2016). The claims did not "claim a new method of virus screening or improvements thereto" and merely claimed use of conventional virus-screening software to carry out the abstract virus-screening idea. Id. And in RecogniCorp, LLC v. Nintendo Co. , we held that claims to encoding and decoding image data were ineligible. 855 F.3d 1322, 1326 (Fed. Cir. 2017). The claims did nothing more than use computers to perform standard encoding and decoding practices. Id. at 1326–27.

The Step 1 "directed to" analysis called for by our cases depends on an accurate characterization of what the claims require and of what the patent asserts to be the claimed advance. The accuracy of those characterizations is crucial to the sound conduct of the inquiries into the problem being addressed and whether the line of specificity of solution has been crossed.

B

Adobe argued to the district court that "the claims are directed to the impermissibly abstract idea of managing access to objects using multiple levels of encryption." J.A. 1845; see J.A. 1857 (heading: "The Asserted Claims of the DCOM Patents Are Directed to the Impermissibly Abstract Idea of Managing Access to Objects Using Multiple Levels of Encryption"); id. ("The asserted claims of the DCOM patents are directed to the abstract idea of managing access to objects using multiple levels of encryption, and fail to provide any specific solution for accomplishing that claimed result."). But that characterization of the two representative claims is materially inaccurate. To arrive at it, Adobe had to disregard elements of the claims at issue that the specification makes clear are important parts of the claimed advance in the combination of elements. It suffices to discuss claim 1. Adobe's characterization of what claim 1 as a whole is directed to does not go beyond what is required simply by the claim term "multi-level ... security," which means " ‘security achieved when encrypted objects are nested within other objects which are also encrypted, possibly within other objects, resulting in multiple layers of encryption.’ " J.A. 1853 (Adobe's motion quoting construction) (citing TecSec I , 731 F.3d at 1345 ). But claim 1 requires more. It goes beyond managing access to objects using multiple levels of encryption, as required by "multilevel ... security." Notably, it expressly requires, as well, accessing an "object-oriented key manager" and specified uses of a "label" as well as encryption for the access management. '702 patent, col. 12, lines 4–16. To disregard those express claim elements is to proceed at "a high level of abstraction" that is "untethered from the claim language" and that "overgeneraliz[es] the claim." Enfish , 822 F.3d at 1337 ; see id. at 1338.

As to the key-manager requirement, we explained in TecSec II that the requirement was important during prosecution. Specifically, rather than deleting the claim limitation after an indefiniteness rejection, TecSec amended the specification to provide the required clarity by adding what now appears at '702 patent, col. 1, line 61, to col. 2, line 4, which provided the definition that this court eventually relied on in construing "object-oriented key manager." TecSec II , 658 F. App'x at 582 (construing the term as "a software component that manages the encryption of an object by performing one or more of the functions of generating, distributing, changing, replacing, storing, checking on, and destroying cryptographic keys"). As to the combination of labeling with the required encryption, the specification makes clear that this is part of the focus of the claimed advance. The specification expressly identifies a deficiency of using only multilevel security through encryption requiring keys at more than one level: that approach, the specification says, "is quite unwieldy, inflexible, and difficult to manage by a security officer or key administrator." '702 patent, col. 2, lines 25–29. The specification then explains that it proposes a solution in which "[a] secure method of labelling files or messages that are sent from a sending user to a receiving user over a network" is used "in addition to cryptographic protection ." Id. , col. 2, lines 30–33 (emphasis added). Adobe has not shown how it could properly disregard these claim requirements in determining "what the patent asserts to be the focus of the claimed advance over the prior art." Solutran , 931 F.3d at 1167–68 (internal quotation marks omitted). The focus of the claimed advance cannot ignore all but the multilevel encryption.

The specification elaborates in a way that simultaneously shows that the claims at issue are directed at solving a problem specific to computer data networks. The patent focuses on allowing for the simultaneous transmission of secure information to a large group of recipients connected to a decentralized network—an important feature of data networks—but without uniform access to all data by all recipients. See '702 patent at col. 11, lines 40–48. The proposed improvement involves, among other things, labeling together with encryption. "Using a secure labelling regimen, a network manager or user can be assured that only those messages meant for a certain person, group of persons, and/or location(s) are in fact received, decrypted, and read by the intended receiver." Id. , col. 2, lines 39–43. "[M]any people within a company may have the key necessary to read a data file" that is encrypted and sent to many terminals, but the sender may not want all such people to read the file. Id. , col. 2, lines 45–51. "By employing a secure labelling technique in addition to encryption , the sender can be assured that people having the correct key to decrypt the message but working at different terminals will not receive or be allowed to access the communication." Id. , col. 2, lines 51–55 (emphasis added).

In light of what the claim language and specification establish, we conclude that the claims are directed to improving a basic function of a computer data-distribution network, namely, network security. See SRI , 930 F.3d at 1304 ; Ancora , 908 F.3d at 1348–49. The patent makes clear that the focus of the claimed advance is on improving such a data network used for broadcasting a file to a large audience, with the improvement assertedly being an efficient way for the sender to permit different parts of the audience to see different parts of the file. '702 patent, col. 2, lines 44–57. While non-computer settings may have security issues addressed by multilevel security, it does not follow that all patents relating to multilevel security are necessarily ineligible for patenting. Here, although the patent involves multilevel security, that does not negate the conclusion that the patent is aimed at solving a particular problem of multicasting computer networks.

By way of comparison, in Uniloc , we held the claims at issue to be directed to solving a problem of reducing communication time by using otherwise-unused space in a particular protocol-based system, 957 F.3d at 1305, 1307–09, even though reducing communication time by using such available blank space (or, generally, reducing resource use by using otherwise-unused available resources) is a goal in many settings. Similarly, in Ancora , we held the claims at issue to be directed to solving a problem presented by particularly easy unauthorized use of software by placing the software in an especially secure computer location, 908 F.3d at 1349, even though placing items in especially secure locations to prevent unauthorized access is a goal in many settings. The patent claims at issue here are directed to improving a data network's basic functioning by enabling secure and efficient transmission to intended recipients when use is made of the basic multicasting functionality of such a data network. As the district court said, the claims address the data-network problem presented by "multiple users in multiple locations accessing information at different security levels from a central repository." J.A. 10.

A second inquiry significant in our case law remains—whether the claims provide a "specific" solution in the way our cases discuss. For similar reasons to those mentioned above, Adobe has not persuaded us that the claims fail to do so, because its arguments ignore what plainly are important aspects of the claims and the focus of the claimed advance in the combination. See , e.g. , J.A. 1857 (arguing that the "asserted claims of the DCOM patents are directed to the abstract idea of managing access to objects using multiple levels of encryption, and fail to provide any specific solution for accomplishing that claimed result").

In these circumstances, we conclude that the district court correctly rejected Adobe's ineligibility challenge as Adobe presented it to that court. C

On appeal, Adobe seeks to modify its definition of what the asserted claims are directed to, now contending that "[t]he representative claims are directed to the abstract idea of managing access to objects using multiple layers of encryption and labels ." Cross-Appellant Principal and Response Br. at 61 (emphasis added). This is a new formulation on appeal. Adobe cites J.A. 1853 and 1857 following the just-quoted sentence, but neither those pages nor any others in Adobe's summary-judgment motion identify the encryption/labeling combination as the abstract idea to which the claims are directed. As recited above, Adobe omitted labeling—as well as the object-oriented key manager—from its characterization of the purported abstract idea in the district court.

For the reasons we have set forth, Adobe's change in its "articulation of the purported abstract idea" is substantively significant. Core Wireless Licensing S.A.R.L. v. LG Electronics, Inc. , 880 F.3d 1356, 1362 n.3 (Fed. Cir. 2018) (declining to decide waiver where the appellant's "articulation of the purported abstract idea was advanced for the first time on appeal"). Adobe's analysis on appeal is still deficient.

When Adobe discusses what is now in its formulation of the asserted abstract idea, it does not meaningfully address the combination. Rather, it asserts the "common-place" character of the individual component techniques generally, see Cross-Appellant Principal and Response Br. at 62 (encryption and labeling); id. at 63 ("[t]he only purported contribution is nesting encrypted objects within other encrypted objects, thereby providing ‘multi-level’ encryption"), or speaks at an even higher level of generality of "encoding, converting, encrypting, and controlling access to data," id. at 66. But that approach is insufficient where, as is true here for the reasons we have explained, it is the combination of techniques that is "what the patent asserts to be the focus of the claimed advance over the prior art." Solutran , 931 F.3d at 1168. Moreover, to the extent, if any, that extrinsic evidence is relevant to the summary-judgment analysis at Step 1, the evidence is one-sided. TecSec submitted an expert declaration asserting that the combination of techniques we have discussed was a specific, unconventional improvement in computer network functionality. J.A. 1960–72; see also J.A. 13 (denying Adobe's motion to strike portions of TecSec's declaration). Adobe has pointed to no evidence of its own that it submitted in support of its § 101 motion.

For those reasons, even apart from its departure from the theory presented in the district court, the analysis Adobe has presented on appeal is insufficient to make the showing required for ineligibility.

IV

For the foregoing reasons, the judgment of the district court is reversed in part and affirmed in part, and the case is remanded for further proceedings on TecSec's claim of inducement of infringement.

Costs awarded to TecSec.

REVERSED IN PART, AFFIRMED IN PART, AND REMANDED