Waterford Township General Employees Retirement System, et al. v. Monolithic Power Systems, Inc., et al.

CourtDistrict Court, W.D. Washington
DecidedMay 6, 2026
Docket2:25-cv-00220
StatusUnknown

This text of Waterford Township General Employees Retirement System, et al. v. Monolithic Power Systems, Inc., et al. (Waterford Township General Employees Retirement System, et al. v. Monolithic Power Systems, Inc., et al.) is published on Counsel Stack Legal Research, covering District Court, W.D. Washington primary law. Counsel Stack provides free access to over 12 million legal documents including statutes, case law, regulations, and constitutions.

Bluebook
Waterford Township General Employees Retirement System, et al. v. Monolithic Power Systems, Inc., et al., (W.D. Wash. 2026).

Opinion

1 2

3 4 5 6 7 UNITED STATES DISTRICT COURT WESTERN DISTRICT OF WASHINGTON 8 AT SEATTLE

9 10 WATERFORD TOWNSHIP CASE NO. C25-0220JLR GENERAL EMPLOYEES 11 RETIREMENT SYSTEM, et al., ORDER 12 Plaintiffs, v. 13

MONOLITHIC POWER SYSTEMS, 14 INC., et al., 15 Defendants. 16

17 I. INTRODUCTION 18 Before the court are Defendants’1 motion to dismiss and request for judicial 19 notice. (MTD (Dkt. # 46); Reply (Dkt. # 55); Request (Dkt. # 47); Request Reply (Dkt. 20 # 55); see also Am. Compl. (Dkt. # 39).) Lead Plaintiff Mirko Dardi opposes the motion. 21

1 Defendants are Michael Hsing and Bernie Blegen (together, the “Executive 22 Defendants”) and Monolithic Power Systems, Inc. (“Monolithic”) (collectively, “Defendants”). 1 (Resp. (Dkt. # 49).) The court has considered the parties’ submissions, the relevant 2 portions of the record, and the governing law. Being fully advised,2 the court GRANTS

3 in part and DENIES in part Defendants’ motion. 4 II. BACKGROUND 5 Mr. Dardi brings this putative securities fraud class action on behalf of investors 6 who purchased or otherwise acquired shares of Monolithic common stock between 7 February 8, 2024, and November 11, 2024, inclusive (the “Proposed Class Period”). (See 8 Am. Compl. ¶¶ 1, 163.) Plaintiff alleges that, during the Proposed Class Period, two of

9 Monolithic’s executive officers, Chief Executive Officer Michael Hsing and Chief 10 Financial Officer Bernie Blegen made “materially false and misleading” statements and 11 “omitted material facts” about the quality and performance of Monolithic’s power 12 management integrated circuits (“PMICs”). (Id. ¶¶ 4, 26-27, 45, 58-75.) PMICs are 13 specialized chips that manage voltage regulation, current control, and thermal protection

14 in power modules. (Id. ¶ 45.) Plaintiff further alleges that Defendants’ conduct 15 artificially inflated the price of Monolithic’s common stock, and, when the true extent of 16 the PMICs’ performance issues came to light, the revelation caused a “precipitous decline 17 in the market value of the Company’s common stock,” resulting in significant losses to 18 the putative class. (Id. ¶ 19.) 3

20 2 Neither party requested oral argument, and the court finds that oral argument would not assist it in deciding the motion. (MTD; Resp); see Local Rules W.D. Wash. LCR 7(b)(4). 3 The court takes judicial notice of Monolithic’s stock price. See Fed. R. Evid. 201(b)(2). 21 On February 8, 2024, at the start of the Proposed Class Period, Monolithic’s stock price closed at $737.07 per share. Monolithic Power Systems, Inc. Common Stock (MPWR) Historical Quotes, 22 NASDAQ, https://www.nasdaq.com/market- 1 Below, the court sets forth the factual background as pleaded by Plaintiff before 2 turning to the relevant procedural history.4

3 A. Monolithic and its Business 4 Monolithic is a publicly traded company headquartered in Kirkland, Washington, 5 which designs and manufactures specialized semiconductor devices. (Am. Compl. ¶¶ 25, 6 30-31.) Its integrated circuits, or chips, are essential for ensuring that electronic devices 7 run efficiently, safely, and with minimal energy loss. (Id. ¶ 31.) Monolithic 8 differentiates itself by using a “monolithic” design philosophy in which it integrates “an

9 entire power system onto a single silicon chip” that is streamlined for size and efficiency. 10 (Id. ¶ 41.) 11 To sell its products, Monolithic uses a cadre of technical sales staff and application 12 engineers to aid prospective customers with the design and development of the 13 customers’ own products. (Id. ¶ 32.) Once a customer selects a product design,

14 Monolithic outsources the production of the chips to specialized third-party 15 manufacturers called “foundries.” (Id.) Monolithic credits this process for its ability to 16 17 activity/stocks/mpwr/historical?page=1&rows_per_page=100&timeline=y5 (last visited May 6, 18 2026). On November 11, 2024, at the close of the Proposed Class Period, Monolithic’s stock price closed at $647.31 per share. Id. On February 4, 2025, when this action commenced, 19 Monolithic’s stock price closed at $656.29 per share. Id. As of the date of this Order, the stock price is $1588.12. Id. 20 4 Defendants requested judicial notice of certain exhibits pursuant to Federal Rule of Evidence 201 and the incorporation doctrine. (See generally Request.) Plaintiff disputes only Exhibits 1-4. (See Resp. at 6 n3.) The court agrees with the parties that Defendants’ Exhibits 21 5-21 are judicially noticeable for the reasons explained in Defendants’ briefing, and therefore takes judicial notice of these exhibits. (See Duffy Decl. (Dkt. # 48) ¶¶ 7-23, Exs. 5-21.) The 22 court addresses the disputed Exhibits 1-4 below. 1 “meet customers’ exacting specifications for quality and performance and promotes 2 future product sales.” (Id.)

3 In an era of increased demand for artificial intelligence (“AI”) applications and an 4 influx of capital to construct data centers for the development of AI technologies, 5 Monolithic thrived and experienced great financial success. (Id. ¶ 35.) Monolithic’s 6 most profitable business segment is its enterprise data segment, which sells PMICs that 7 power AI technology and contributes 32.5% of Monolithic’s revenue. (Id. ¶¶ 47-48.) 8 Specifically, the enterprise data end market relates to “data-center [sic] and cloud

9 infrastructure markets, in particular, power solutions for servers, storage, networking, and 10 AI computing hardware.” (Id. ¶ 48.) Enterprise data customers are primarily technology 11 companies that engage Monolithic to design power management modules for use in 12 graphic processing units (“GPUs”) that run and operate AI data servers. (Id. ¶ 33.) 13 During the Proposed Class Period, Nvidia Corporation (“Nvidia”) was

14 Monolithic’s most important enterprise data customer. (Id. ¶ 48.) Nvidia is the world’s 15 leading supplier of GPUs, “which are specialized electronic circuits built for parallel 16 processing—performing many mathematical operations simultaneously.” (Id. ¶ 49.) 17 Nvidia’s GPUs aid with AI acceleration, deep learning, and high-performance 18 computing. (Id.) Monolithic did not sell directly to Nvidia. (Id. ¶ 54.) Rather, Nvidia

19 purchased Monolithic’s PMICs from a network of third-party distributors and value- 20 added resellers or directly from contract manufacturers or board assemblers. (Id.) These 21 retailers integrated Monolithic’s PMICs into Nvidia’s GPUs and other machinery. (Id.) 22 1 In 2022, Nvidia launched the Hopper GPU to construct the next wave of AI data 2 centers. (Id. ¶ 34.) Monolithic supplied the PMICs for Nvidia’s Hopper architecture –

3 the H100 – which was designed for “large-scale computing applications and excelled in 4 transformer-based AI models, large language models (LLMs), and scientific workloads.” 5 (Id. ¶ 50.) AI servers widely used the H100 through 2024. (Id.) As the sole supplier of 6 the PMICs used in the H100, Monolithic experienced exponential growth in its enterprise 7 data business as a result of increased customer demand for its AI power solutions. (Id. 8 ¶¶ 35, 38; id. ¶ 38 (alleging that Monolithic’s quarterly enterprise data revenue increased

9 290% year-over-year, from $48 million to $187 million).) 10 In 2024, Nvidia released its Blackwell architecture as the successor to the Hopper 11 GPU. (Id. ¶ 55.) Blackwell delivers better performance and energy efficiency than 12 Hopper. (Id. ¶ 56 (“Blackwell delivers 2.5x performance increase over Hopper while 13 being 25x more energy efficient and six times faster than Hopper for large datasets.”).)

Free access — add to your briefcase to read the full text and ask questions with AI

Related

Foman v. Davis
371 U.S. 178 (Supreme Court, 1962)
Tellabs, Inc. v. Makor Issues & Rights, Ltd.
551 U.S. 308 (Supreme Court, 2007)
Uttecht v. Brown
551 U.S. 1 (Supreme Court, 2007)
Watters v. Wachovia Bank, N. A.
550 U.S. 1 (Supreme Court, 2007)
United States v. Laurienti
611 F.3d 530 (Ninth Circuit, 2010)
United States v. Sepulveda
15 F.3d 1161 (First Circuit, 1993)
In Re Apple Computer Securities Litigation
886 F.2d 1109 (Ninth Circuit, 1989)
Vess v. Ciba-Geigy Corp. USA
317 F.3d 1097 (Ninth Circuit, 2003)
Zucco Partners, LLC v. Digimarc Corp.
552 F.3d 981 (Ninth Circuit, 2009)
Metzler Investment GMBH v. Corinthian Colleges, Inc.
540 F.3d 1049 (Ninth Circuit, 2008)
Von Saher v. Norton Simon Museum of Art at Pasadena
592 F.3d 954 (Ninth Circuit, 2010)
In Re Gilead Sciences Securities Litigation
536 F.3d 1049 (Ninth Circuit, 2008)
In Re NUVELO, INC, SECURITIES LITIGATION
668 F. Supp. 2d 1217 (N.D. California, 2009)

Cite This Page — Counsel Stack

Bluebook (online)
Waterford Township General Employees Retirement System, et al. v. Monolithic Power Systems, Inc., et al., Counsel Stack Legal Research, https://law.counselstack.com/opinion/waterford-township-general-employees-retirement-system-et-al-v-wawd-2026.