MEMORANDUM ORDER

CHRISTOPHER J. BURKE, Magistrate Judge.

Presently pending before the Court is Defendant Intel Corporation's ("Defendant") request that the Court strike what Defendant refers to as "newly identified product families" (the "products at issue") from Plaintiff VLSI Technology, LLC's ("Plaintiff") amended infringement contentions, which Plaintiff served on July 24, 2019. (D.I. 320 at 1) For the reasons set forth below, the Court concludes that Plaintiff untimely identified the products at issue as infringing products, but reserves decision on whether to exclude such products from the case pending further briefing.

I. BACKGROUND

On October 7, 2019, the parties advised the Court that they were seeking resolution of several discovery disputes, including the instant dispute; on October 21, 2019, the parties filed a joint motion ("Motion") in that regard.1 (D.I. 298; D.I. 345) The Court thereafter considered the parties' initial briefing with regard to all of these disputes, (D.I. 319-20; D.I. 325-26), and held telephonic oral argument on the Motion, (D.I. 456). On October 21 and 23, 2019, the Court issued three separate Oral Orders addressing the remaining disputes raised in the Motion other than the instant dispute. (D.I. 338; D.I. 343; D.I. 351)

With regard to the instant dispute, Defendant challenged as untimely Plaintiff's amended infringement contentions relating to United States Patent Nos. 8,081,026 (the "'026 patent"), 7,246,027 (the "'027 patent"), 7,523,331 (the "'331 patent") and 7,247,552 (the "'552 patent"). (D.I. 320 at 2; id., ex. 1) In a Memorandum Order issued on October 25, 2019, the Court resolved a portion of this dispute. (D.I. 363) Defendant's position was that Plaintiff's amended infringement contentions added new, previously-unaccused "product families" to the case. (D.I. 320 at 1-2) Plaintiff disagreed, arguing that it had simply accused "sub-families within the accused product families already identified." (D.I. 325 at 1 (emphasis in original) (cited in D.I. 363 at ¶ 1)) The Court agreed with Plaintiff that with regard to issues of infringement and invalidity, discovery relating to Defendant's products has proceeded on a product-family-by-product-family basis. (D.I. 363 at ¶ 2) Indeed, the Court explained that Defendant had previously conceded that identifying products in this way (i.e., on a product-family-by-product-family basis) was appropriate, because (at least for purposes of this case), products that are a part of each Defendant product family share the same material underlying product architecture. (Id. at ¶ 2 (citations omitted)) And the Court also noted that in June 2019, the District Court had ordered that Plaintiff and Defendant must specifically identify a product family as to their respective infringement and invalidity contentions. (Id.; D.I. 455 at 38, 44-45)

From there, the Court reasoned that "so long as there is sufficient record information to allow [it] to conclude that what Plaintiff has done in its amended infringement contentions is to merely identify products that fall within an already-identified Defendant product family (i.e., "sub-families")—as opposed to identifying new Defendant product families—then such amendment would not be untimely or otherwise problematic." (Id. at 3) As to the '026 patent and the '027 patent, the Court had before it sufficient information to determine that the allegedly newly-added product families at issue (i.e., Amber Lake Y and Whiskey Lake) were not considered by Defendant to be new or different product families; instead, the record showed that Defendant considered them to be product sub-families within a previously-identified product family (i.e., the Skylake family). (Id. at 3) And so, the Court denied Defendant's request to strike the inclusion of those sub-families from Plaintiff's amended infringement contentions. (Id.)

But as to the '331 patent and the '552 patent, the Court explained that it did not then have a sufficient record to determine whether the allegedly newly-added product families are actually subfamilies that are subsumed within previously disclosed product families. (Id. at 4) Thus, it ordered the parties to provide supplemental letter briefing as to that issue. (Id.). The parties filed these supplemental letter briefs on November 1, 2019. (D.I. 370-71)

II. STANDARD OF REVIEW

Paragraph 3 of the Scheduling Order in this case incorporates certain provisions of the Court's Default Standard for Discovery, Including Discovery of Electronically Stored Information ("Default Standard"). (D.I. 40 at 1-2) And the Scheduling Order in turn required Plaintiff to "specifically identify" the accused products at issue in the case by November 12, 2018, and to provide initial infringement contentions on January 23, 2019. (Id. at 2; Default Standard at 4)2

Federal Rule of Civil Procedure 26(e) states that a party who has made a disclosure under Federal Rule of Civil Procedure 26(a), or who has responded to an interrogatory, request for production or request for admission, "must supplement or correct its disclosure or response" in a "timely manner" if it "learns that in some material respect the disclosure or response is incomplete or incorrect, and if the additional or corrective information has not otherwise been made known to the other parties during the discovery process or in writing[.]" Fed. R. Civ. P. 26(e)(1).3 Courts consider a number of factors in determining whether a party has breached its duty to timely amend a discovery disclosure or response under this rule, including: (1) whether there was a prior response; (2) whether the response became materially incorrect or incomplete; (3) whether the party knew that the response was incomplete; and (4) whether the corrective information was otherwise made known to the other party through the discovery process or in writing. Lambda Optical Sols., LLC v. Alcatel-Lucent USA Inc., Civil Action No. 10-487-RGA-CJB, 2013 WL 1776104, at *2 (D. Del. Apr. 17, 2013) (citing cases). The focus under these factors is whether a party provided adequate notice of its legal contentions and their corresponding evidentiary bases. Id.

If a party fails to timely provide information pursuant to Rule 26, that party is "not allowed to use that information ... to supply evidence .... unless the failure was substantially justified or is harmless." Fed. R. Civ. P. 37(c)(1). "It is left to the trial court's discretion to determine whether a party provides substantial justification for their delay or if the delay is harmless." Amgen Inc. v. Aurobindo Pharma Ltd., Civil Action No. 16-853-GMS, 2018 WL 1061369, at *4 (D. Del. Feb. 27, 2018), reargument denied sub nom. Amgen Inc. v. Amneal Pharms. LLC, Civ. No. 16-853-GMS, 2018 WL 1885664 (D. Del. Apr. 19, 2018). In exercising this discretion, the court should consider what are referred to as the "Pennypack factors": (1) the importance of the information withheld; (2) the prejudice or surprise in fact to the opposing party, (3) the likelihood of disruption of the trial; (4) the possibility of curing the prejudice; (5) the explanation for the failure to disclose; and (6) the presence of bad faith or willfulness in not disclosing the evidence. Lambda, 2013 WL 1776104, at *2 & n.3 (citing cases); see also Amgen Inc., 2018 WL 1061369, at *4.

III. DISCUSSION

Below, the Court will first assess whether Plaintiff's amended infringement contentions were timely. Thereafter, it will address the issue of whether, if the contentions were untimely, evidence relating to them should be excluded from this case.

A. Were Plaintiff's Amended Infringement Contentions as to the '331 Patent and the '552 Patent Timely?

In asserting that its July 2019 amended infringement contentions were timely, Plaintiff makes a few different arguments. The Court will assess those arguments in turn.

1. Are The Products at Issue Simply "Sub-family" Members of a Defendant Product Family that Plaintiff Had Already Specifically Accused of Infringing These Patents?

First, Plaintiff argues that each of the products at issue simply amount to "sub-families" of Defendant product families that Plaintiff specifically accused of infringement earlier in the case. (D.I. 370 at 1-3) In other words, Plaintiff is asserting that since this is so, then when it earlier timely accused the Defendant product "families" that contain these "sub-families," it had then necessarily timely accused those sub-families of infringement as well. In the Court's view, however, the record does not support Plaintiff's assertion.

The Court first assesses the products at issue relating to the '331 patent. In in its January 2019 initial infringement contentions, Plaintiff specifically accused certain of Defendant's products both by referencing the product's Platform Controller Hub ("PCH" or "chipset") version and by referencing the product's firmware; as to the product's firmware, Plaintiff identified that firmware by reference to its Management Engine ("ME") or Converged Security and Management Engine ("CSME") version number. (D.I. 370 at 1 & ex. 1) At issue now are certain PCH/chipset versions that Plaintiff identified for the first time in its amended contentions in July 2019 (e.g., the Tylersburg Chipset, Patsburg Chipset, Wellsburg Chipset, Coffee Lake Chipset and Whiskey Lake Chipset). (D.I. 370 at 1 & ex. 2) Each of these PCH/chipset versions are said by Plaintiff to utilize an ME or CSME firmware version that was previously identified in the January 2019 initial infringement contentions (i.e., an ME or CSME firmware version that was there listed as being associated with a different PCH/chipset). Plaintiff's argument is that: (1) since the ME/CSME versions are the "product families at issue for the '331 patent" and (2) since it had identified those "product families" as accused back in January 2019, then (3) the fact that it is now newly accusing PCH/chipset "sub-families" of those previously-identified ME/CSME version "families" is not problematic from a notice perspective.

The problem for Plaintiff is that there is no evidence that Defendant actually considers the products at issue to be members of a "family" defined by its ME/CSME firmware component—and there is some evidence that it does not. The Court so concludes partly because Defendant has asserted in its briefing that it "does not itself group products into `families' based on which version of `ME' they contain[,]" since many of its products allegedly "share firmware components while also having fundamentally different architecture" and because those products use ME in different ways. (D.I. 371 at 3) But the Court also takes note that even Plaintiff—as late as in its July 2019 amended contentions—has referred to the PCH/chipset version as the relevant "Product Family" with regard to products accused of infringing this patent. (D.I. 370, ex. 2) This suggests that as late as July 2019, even Plaintiff did not believe that the relevant product families at issue here were defined by reference to ME/CSME versions.

The Court's analysis of the products at issue regarding the '552 patent is similar. In its January 2019 initial infringement contentions, Plaintiff accused all of Defendant's products "fabricated using reinforcing metal lines running in bond pad regions in the manner described below[.]" (D.I. 320, ex. 1 at 8) It also specifically accused one of Defendant's products simply by referencing the "Processor" used therein—"Haswell"—and the processor's "Apparent Release Year"—2013. (D.I. 320, ex. 1 at 8-11) Then on May 20, 2019, Plaintiff informed Defendant by letter that it was also specifically accusing six additional "Intel Products" of infringing the '552 patent ("Broadwell, Skylake, Kaby Lake, Coffee Lake, Cannon Lake and Whiskey Lake"). (D.I. 370, ex. 13)4 Yet in its July 2019 amended infringement contentions, Plaintiff changed course. There, it explicitly defined the relevant accused "Product Family" by reference to the process node on which the products were manufactured (i.e., the 22 nanometer ("nm"), 14 nm and 10 nm process nodes), and then went on to list 47 different newly-identified purported "Exemplary Sub-Families" of those purported process node "Product Famil[ies]" (D.I. 320, ex. 1 at 8-11; D.I. 370 at 2-3) Plaintiff now argues that with these July 2019 amended contentions, it was simply identifying "process [node] families" that had already been previously accused in the case. (D.I. 370 at 3)5

Again, however, there is no evidence that Defendant actually considers the 47 newly-identified purported "Sub-Families" at issue to actually be a part of a "product family" that is defined by the process node on which the products were manufactured. And there is evidence that it does not. To that end, there is evidence of record that Defendant refers to its "product families" by reference to their processor (i.e., the "Haswell family" of products). (D.I. 358, ex. A at 67 (Defendant's deposition witness acknowledging, in response to a question, the existence of the "Haswell family of products")) Plaintiff itself appeared to believe that the relevant product families as to this patent were defined by reference to the "Processor" included therein— as evinced by its January 2019 initial contentions and its May 2019 letter. (D.I. 320, ex. 1 at 8-11; D.I. 370, ex. 13)6 Finally, in its briefing, Defendant asserts that it "does not group products into `families' based on process node" because "[p]roducts that share a process node may vary drastically in kind and in characteristics." (D.I. 371 at 2)

In sum, Defendant has conceded (and the District Court has also concluded) that if Plaintiff had previously identified (i.e., prior to July 2019) a Defendant "product family" and accused that product family of infringing either of these two patents, then such an allegation would have sufficed to timely accuse any of the "sub-family" members of those product families. (D.I. 455 at 41-43, 45) But here, Plaintiff cannot make use of this concession to show that the accused products at issue were timely accused. This is because the evidence does not show that the products at issue actually are considered to be sub-family members of a previously-accused Defendant product family. And so if Plaintiff wants to demonstrate the timeliness of its July 2019 amended contentions, it will have to try another argument.

2. Even if the Accused Products at Issue are Not Part of a Defendant Product Family that Plaintiff Had Specifically Accused of Infringement, Did Plaintiff, Prior to Filing its July 2019 Amended Contentions, Otherwise (i.e., Through the Discovery Process or In Writing) Do Enough to Put Defendant on Notice that These Products Were Being Accused of Infringement?

Another way that Plaintiff can demonstrate that the accused products at issue here were timely accused is by demonstrating that, prior to July 2019, this fact was "otherwise ... made known to [Defendant] through the discovery process or in writing[.]" Fed. R. Civ. P. 26(e)(1). Here, Plaintiff argues that it did so. That is, Plaintiff asserts that it otherwise gave Defendant notice earlier in the discovery process that it was accusing all such products: (1) employing certain ME/CSME version numbers or (2) having a 22 nm, 14 nm or 10 nm process node designation. (D.I. 370 at 2-4)

The Court, however, disagrees that this argument helps Plaintiff. In explaining why, it turns to its prior decision in Invensas Corp. v. Renesas Electronics Corp., 287 F.R.D. 273 (D. Del. 2012).

In Invensas, the Court summarized the state of the law regarding the extent to which a patentee has a right to discovery with regard to an accused infringer's products. It noted that this Court's Default Standard, which is often (as it was here) incorporated into patent Scheduling Orders, provides one easy mechanism for a patentee to confirm that a set of the accused infringer's products are properly accused. Invensas, 287 F.R.D. at 283. That is, the patentee can specifically accuse such products by name, early in the case. Id.

But what if the patentee believes that the accused infringer makes additional products that likely infringe the patent-in-suit, but does not know the identity or name of such products? Can the patentee properly get access to discovery on such non-specifically-accused products? In Invensas, the Court attempted to answer this question. The Court concluded that on the one hand, the fact that a patentee has not yet been able to identify an alleged infringer's product by name does not mean that discovery as to such products is de facto irrelevant to a case. Id. at 281. But on the other hand, the Court explained that such discovery is not de facto relevant simply because the patentee wishes to obtain it. Id. at 281-82. Instead, the Court noted that if a plaintiff seeks discovery on non-specifically-accused products, and if a defendant contests whether such discovery is appropriate, the plaintiff has to do some work. It has to demonstrate to a court that such discovery is warranted, a decision that will likely turn on three factors: "(1) as to relevance, the specificity with which the plaintiff has articulated how the unaccused products are relevant to its existing claims of infringement ... (2) whether the plaintiff had the ability to identify such products via publicly available information prior to the request and (3) the nature of the burden on defendant(s) to produce the type of discovery sought." Id. at 282.

In this case, with regard to the products at issue, Plaintiff obviously did not identify them by name (pursuant to the Default Standard) as part of its initial patent disclosures. And Plaintiff must have known that its initial disclosures were materially incorrect or incomplete at the time they were filed, or soon after. After all, when Plaintiff made those disclosures, it believed that other of Defendant's products were also infringing the patents-at-issue. (D.I. 370 at 3-4)

Now, to be sure, Plaintiff did make some attempts early in the discovery process to try to obtain discovery on these non-specifically-accused products. For example, with regard to the '331 patent, in its January 2019 initial infringement contentions, Plaintiff stated that it wished to accuse "all Intel products that include" the particular ME/CSME firmware versions that were part of the products at issue. (D.I. 370, ex. 1 at 1; see also D.I. 370 at 1 (Plaintiff noting that its July 2019 amended contentions "did not add any new version[s] of the ME/CSME")) With regard to the '552 patent, as early as December 2018 Plaintiff requested from Defendant documents regarding the "P1270 manufacturing process" (an internal Defendant code name for the 22 nm process), asserting that such documents were relevant to its infringement allegations regarding this patent. (D.I. 370 at 3; id., ex. 7 at 2) Plaintiff also notes that in late 2018, it served Requests for Production on Defendant seeking discovery as to infringement of this patent on any products, inter alia, that use "Integrated Circuit Packaging[.]" (Id., ex. 22-23) In this way, Plaintiff is suggesting that it "made known" that it was going to be accusing the products at issue and others like them of infringement.

Yet a key point here is that Defendant did not agree that these broad requests for product discovery were sufficiently detailed, such that it was properly on notice of what accused products (beyond those that had been specified by name in Plaintiff's initial infringement contentions) were really at issue in the case. (See, e.g., id., ex. 9 at 8; id., ex. 10 at 11) Instead, Defendant argued that Plaintiff's requests amounted to wrongly (and vaguely) accusing "like everything Intel does" of infringement. (D.I. 455 at 59; see also D.I. 371 at 2-3) And so Defendant did not agree to search for or produce such discovery. (Id.) In other words, Defendant was contesting that by referencing these non-specifically-accused products in this way, Plaintiff had done enough to "provide[] adequate notice of its legal contentions and their corresponding evidentiary bases." Lambda, 2013 WL 1776104, at *2.

Thus, at that stage of the case—if Plaintiff wanted to be assured that it had done enough to provide the required notice, and that it had demonstrated the relevance of such products—then it needed to take further action. It needed to make a sufficient record to persuade a judge that it had met the Invensas test (or had otherwise complied with the Federal Rules of Civil Procedure) as to such non-specifically-accused products.

But for whatever reason, Plaintiff did not bring that dispute to the District Court until June 2019. And ultimately, after receiving briefing on the issue, (D.I. 173 at 2-3; D.I. 178 at 2-3), and hearing argument, (D.I. 455 at 54-61), the District Court agreed with Defendant. It found that "the definition that's utilized [by Plaintiff in seeking discovery on non-specifically-accused products] is far too broad" and determined that discovery need only be provided as to specifically-identified Defendant product families. (Id. at 61; see also D.I. 370 at 2 (Plaintiff noting that "at the June 18 hearing, [the District Court] limit[ed] accused products in this case to those identified by family")) In other words, the District Court determined that Plaintiff was not legally entitled to discovery on products that Plaintiff had not identified at least by reference to Defendant product family name.

In light of this, the Court cannot conclude that, prior to July 2019, Plaintiff had "otherwise made known to [Defendant] through the discovery process or in writing" that it was properly accusing the products at issue. Fed. R. Civ. P. 26(e). How could the Court so conclude, when the District Court has already determined that Plaintiff's efforts in this regard were insufficient—and that such products (as of the June 2019 discovery conference) were not properly in the case? Thus, this argument cannot be a basis for concluding that Plaintiff timely identified the products at issue.

3. Was Plaintiff's Specific Identification of the Products at Issue in the July 2019 Amended Contentions Otherwise Timely?

The Court lastly considers whether Plaintiff's July 24, 2019 amended contentions, which for the first time specifically identified the products at issue by relevant family name, were timely.

On the one hand, there is no deadline in the operative Scheduling Order for the submission of final infringement contentions, (see D.I. 40 at ¶¶ 3-4), and the Default Standard expressly permits supplementation. Default Standard at 4 n.3. And although the July 2019 amended contentions were served over eight months from the date for first specifically identifying accused products and six months from the date that initial infringement contentions were due, they were still provided to Defendant in advance of the fact discovery deadline.7 Lambda, 2013 WL 1776104, at *4 (noting that as a general matter, if there is no deadline for final amendment of infringement contentions, and if the contentions at issue are provided well before the close of fact discovery, this can weigh in favor of a finding that amended contentions were not untimely) (citing cases).

On the other hand, for months prior to July 2019, Plaintiff knew that there was a dispute brewing over whether such non-specifically-accused products were properly in the case. Moreover, and most significantly, it appears that Plaintiff could have acted earlier than July 2019 to learn about the identity of the products at issue.

For example, as to the products at issue relating to the '331 patent, the Court does not see why Plaintiff could not have identified them by family name months earlier. After the District Court's June 2019 discovery ruling, Plaintiff says that it "created a script that data-mined [Defendant's] website http://ark.intel.com to obtain such information" as to "which PCHs/chipsets include the accused ME/CSME versions[.]" (D.I. 370 at 2) Left unsaid is why Plaintiff could not have taken this path long before. (D.I. 371 at 4 (Defendant asserting that "[t]he fact that VLSI did not do so before July is a result of its delay, not because it could not have done so"); see also D.I. 456 at 71-75)

As for the products at issue regarding the '552 patent, Plaintiff explains that it was originally able to accuse Defendant's Haswell project (a 22 nm process) of infringement after performing a "product tear-down[.]" (D.I. 370 at 3) In light of that, Defendant argues that Plaintiff could have performed tear-downs of the other products at issue months ago. (D.I. 371 at 4 & n.3) Plaintiff appears to respond by asserting that such tear-downs are "very expensive and laborious[,]" (D.I. 370 at 3), but it provided no other information about how prohibitively expensive or time-intensive it would have been to take such actions.8 Cf. Invensas, 287 F.R.D. at 285 (noting that to the extent that a patentee "[could not] simply go into a store, easily locate Defendants' [now-accused products] purchase them, tear them down and analyze them[,]" this helps credibly explain why the patentee did not earlier accuse such products of infringement by name). And back in May 2019, Plaintiff was also able to accuse by name six other product families relating to 14 nm and 10 nm process chips. (D.I. 370 at 4; id. ex. 19 at 1) If Plaintiff, through its investigation, was able to identify these six new product families in May, why could it not then have identified all of the products at issue later named in its amended disclosures?

Because it appears to the Court that Plaintiff could have reasonably taken efforts to accuse the products at issue well before it did so, the Court concludes that Plaintiff's July 2019 amended contentions were untimely.

B. The Pennypack Factors

Despite the Court having concluded that Plaintiff's amended contentions were untimely, this does not mean that they necessarily should be stricken or that evidence relating to them may not be used in this case. The United States Court of Appeals for the Third Circuit has noted that exclusion of critical evidence is an "extreme sanction" that is not normally imposed absent a showing of willful deception or flagrant disregard of a court order by the proponent of the evidence. In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 791-92 (3d Cir. 1994) (internal quotation marks and citations omitted); Int'l Bus. Machs. Corp. v. Priceline Grp., Inc., 271 F.Supp.3d 667, 694 (D. Del. Sept. 18, 2017); Lambda, 2013 WL 1776104, at *2. The Court will need to apply the Pennypack factors in order to determine whether to exclude the evidence.

Yet the Court does not have enough information to reasonably apply those factors. The parties did not address the Pennypack factors by name in their supplemental briefs. Nor did they address the substance of most of the factors in that briefing. And the Court also has unanswered questions about the application of those factors here, including: (1) Exactly how much time and effort would it take to produce documents relating to the products at issue?9; (2) Why would (or would not) the November 2019 trial date be threatened if the products at issue were in the case?; (3) If the products at issue were excluded, how would legal issues of claim preclusion or issue preclusion likely affect Plaintiff's ability to assert infringement of the patents as to those products in a future case?; (4) What impact (economic or otherwise) does the products' inclusion have on the damages issues in the case?; and (5) Why did it take Defendant from July 24, 2019 until the filing of its Motion in October 2019 to move to exclude the products at issue?

Thus, by no later than December 20, 2019, each party shall file a supplemental letter brief, of no more than four single-spaced pages, addressing how the Pennypack factors apply to this matter.

IV. CONCLUSION

For the reasons set out above, the Court concludes that Plaintiff's July 2019 amended infringement contentions were untimely pursuant to Rule 26(e). The Court reserves judgment, however, on whether to grant Defendant's Motion, pending a review of further supplemental briefing regarding the application of Rule 37 and the Pennypack factors.

Because this Memorandum Order may contain confidential information, it has been released under seal, pending review by the parties to allow them to submit a single, jointly proposed, redacted version (if necessary) of the document. Any such redacted version shall be submitted by no later than December 18, 2019 for review by the Court, along with a motion for redaction that includes a clear, factually detailed explanation as to why disclosure of any proposed redacted material would "work a clearly defined and serious injury to the party seeking closure." Pansy v. Borough of Stroudsburg, 23 F.3d 772, 786 (3d Cir. 1994) (internal quotation marks and citation omitted). The Court will subsequently issue a publicly-available version of its Memorandum Order.