California Public Employees v. Palantir Technologies Inc. — Entry #32: [11224750] Appellant/Petitioner's reply brief filed by California Public Employees' Retirement System and Shijun Liu
Case: California Public Employees v. Palantir Technologies Inc. ca10 · 25-1178
filed May 06, 2025
What this document is
Docket entry #32 · filed October 21, 2025
[11224750] Appellant/Petitioner's reply brief filed by California Public Employees' Retirement System and Shijun Liu. Served on 10/21/2025. Manner of Service: email. Word/page count: 6482. This pleading complies with all required privacy and virus certifications: Yes. [25-1178] JDD [Entered: 10/21/2025 03:55 PM]
Who is involved
- ALEXANDER C. KARP
- ALEXANDER MOORE
- ALEXANDRA SCHIFF
- CALIFORNIA PUBLIC EMPLOYEES' RETIREMENT SYSTEM
- DAVID GLAZER
- KEVIN KAWASAKI
- PALANTIR TECHNOLOGIES INC.
- PETER THIEL
- SHIJUN LIU, individually and on behalf of all others similarly situated trustee Liu Family Trust 2019
- SHYAM SANKAR
- SPENCER RASCOFF
- STEPHEN COHEN
- WILLIAM HO
Why we have it
We follow this case because it names a company we track, although that company is not a party:
- Riot: its former name “AspenBio Pharma, Inc.” appears in a filing in this case.
…....................... 44 Wolfe v. Aspenbio Pharma, Inc., 587 F. App’x 493 (10th Cir. 2014)…
A free copy from the RECAP archive of federal court filings (mirrored at the Internet Archive), retrieved November 15, 2026. Federal court filings are public records.
Document text
40 page(s), 68,017 characters, converted from the PDF's text layer · plain text.
Full text
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 1
No. 25-1178
IN THE UNITED STATES COURT OF APPEALS
FOR THE TENTH CIRCUIT
CALIFORNIA PUBLIC EMPLOYEES’ RETIREMENT SYSTEM, et al.,
Plaintiffs/Appellants,
vs.
PALANTIR TECHNOLOGIES INC., et al.,
Defendants/Appellees.
Appeal from a Decision of the United States District Court
for the District of Colorado, Case No. 1:22-cv-02384-GPG-SBP
Honorable Gordon P. Gallagher, U.S. District Judge
PLAINTIFFS-APPELLANTS’ REPLY BRIEF
ROBBINS GELLER RUDMAN
& DOWD LLP
DARREN J. ROBBINS
JOSEPH D. DALEY
LUKE O. BROOKS
J. MARCO JANOSKI GRAY
ERIKA L. OLIVER
655 West Broadway, Suite 1900
San Diego, CA 92101
Telephone: 619/231-1058
619/231-7423 (fax)
Counsel for Plaintiffs/Appellants
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 2
TABLE OF CONTENTS
Page
I. ARGUMENT ...................................................................................................1
A. Plaintiffs’ allegations confirm that Defendants acted with
scienter in connection with their scheme to defraud, false-and-
misleading statements, and insider selling. ........................................... 1
1. Defendants’ arguments revolve around a specious theme
and several false premises. .......................................................... 1
2. Pluralsight confirms Defendants’ billions of dollars in
insider sales at inflated prices bolster a strong inference
of scienter. ...................................................................................4
3. Defendants’ construction and utilization of the SPAC
scheme was deliberate misconduct. ..........................................11
4. Defendants’ conceded access to truthful information
bolsters a strong inference of scienter.......................................18
5. Defendants’ trumpeted “strong visibility” into Palantir’s
revenue growth further establishes their scienter. ....................23
6. Defendants’ deliberate manipulation of key reporting
metrics cannot be explained away. ...........................................25
B. Defendants haven’t satisfactorily defended the district court’s
failure to address the deliberate actions comprising their
scheme. ................................................................................................29
II. CONCLUSION..............................................................................................32
CERTIFICATE OF COMPLIANCE WITH TYPE-VOLUME LIMIT.................. 33
DECLARATION OF SERVICE .............................................................................34
-i-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 3
TABLE OF AUTHORITIES
Page
CASES
Adams v. Kinder-Morgan, Inc.,
340 F.3d 1083 (10th Cir. 2003) ......................................................................4, 21
Aldridge v. A.T. Cross Corp.,
284 F.3d 72 (1st Cir. 2002) ...........................................................................15, 16
Ashcroft v. Iqbal,
556 U.S. 662 (2009) ....................................................................................1, 2, 13
Bell Atl. Corp. v. Twombly,
550 U.S. 544 (2007) ....................................................................................1, 2, 31
Cent. Laborers’ Pension Fund v. Karp,
2025 WL 1213104 (Del. Ch. Apr. 25, 2025) ........................................................ 3
City of Warren Police & Fire Ret. Sys. v. Prudential Fin., Inc.,
70 F.4th 668 (3d Cir. 2023) ................................................................................20
Clem v. Lomeli,
566 F.3d 1177 (9th Cir. 2009) .............................................................................. 4
Clinton v. Sec. Benefit Life Ins. Co.,
63 F.4th 1264 (10th Cir. 2023) .................................................................2, 13, 30
Dirks v. SEC,
463 U.S. 646 (1983) ............................................................................................12
Employees’ Ret. Sys. of Gov’t of the Virgin Is. v. Blanford,
794 F.3d 297 (2d Cir. 2015) ...............................................................................21
Frank v. Dana Corp.,
547 F.3d 564 (6th Cir. 2008) ..............................................................................31
Goldman v. Belden,
754 F.2d 1059 (2d Cir. 1985) ...............................................................................7
- ii -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 4
Page
Harris v. Union Elec. Co.,
787 F.2d 355 (8th Cir. 1986) ................................................................................ 3
Helwig v. Vencor, Inc.,
251 F.3d 540 (6th Cir. 2001) (en banc) ................................................7, 8, 16, 17
Herman & MacLean v. Huddleston,
459 U.S. 375 (1983) ............................................................................................16
In re Alphabet, Inc. Sec. Litig.,
1 F.4th 687 (9th Cir. 2021) .................................................................................26
In re Daou Sys.,
411 F.3d 1006 (9th Cir. 2005) ............................................................................21
In re Genius Brands Int’l, Inc. Sec. Litig.,
97 F.4th 1171 (9th Cir. 2024) .............................................................................29
In re Grand Casinos Sec. Litig. Inc.,
988 F. Supp. 1273 (D. Minn. 1997)....................................................................17
Ind. Pub. Ret. Sys. v. Pluralsight, Inc.,
45 F.4th 1236 (10th Cir. 2022) ............................ 2, 4-9, 11, 15, 20, 23-25, 29, 31
Johnson v. Reyna,
57 F.4th 769 (10th Cir. 2023) .............................................................................30
Kramer v. Time Warner, Inc.,
937 F.2d 767 (2d Cir. 1991) ................................................................................. 6
Lormand v. US Unwired, Inc.,
565 F.3d 228 (5th Cir. 2009) ..............................................................................21
Lovelace v. Software Spectrum,
78 F.3d 1015 (5th Cir. 1996) ................................................................................6
- iii -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 5
Page
Meitav Dash Provident Funds & Pension Ltd. v. Spirit AeroSystems
Holdings, Inc.,
79 F.4th 1209 (10th Cir. 2023) .....................................................................18, 22
Omnicare, Inc. v. Laborers Dist. Council Constr. Indus. Pension Fund,
575 U.S. 175 (2015) ............................................................................................20
Pirraglia v. Novell, Inc.,
339 F.3d 1182 (10th Cir. 2003) ..............................................................16, 26, 28
Plotkin v. IP Axess Inc.,
407 F.3d 690 (5th Cir. 2005) ..............................................................................25
Reese v. Malone,
747 F.3d 557 (9th Cir. 2014), overruled in part on other grounds
by City of Dearborn Heights Act 345 Police & Fire Ret. Sys. v. Align Tech.,
Inc., 856 F.3d 605 (9th Cir. 2017) ......................................................................17
Ronconi v. Larkin,
253 F.3d 423 (9th Cir. 2001), abrogation recognized by Cai v. Eargo,
Inc., 2025 WL 66041 (9th Cir. Jan. 10, 2025) ................................................9, 10
Schleicher v. Wendt,
618 F.3d 679 (7th Cir. 2010) ................................................................................ 3
Smallen v. W. Union Co.,
950 F.3d 1297 (10th Cir. 2020) ................................................................9, 10, 11
Tellabs, Inc. v. Makor Issues & Rts., Ltd.,
551 U.S. 308 (2007) ................................................. 1-2, 13, 17, 20, 23-24, 29-31
Va. Bankshares, Inc. v. Sandberg,
501 U.S. 1083 (1991) ..........................................................................................26
- iv -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 6
Page
STATUTES, RULES, AND REGULATIONS
15 U.S.C.
§77k.....................................................................................................................20
Delaware Chancery Court
Rule 23.1 ...............................................................................................................3
17 C.F.R.
§240.10b-5 ..........................................................................................................20
§240.10b5-1 ......................................................................................................5, 6
-v-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 7
I. ARGUMENT
Plaintiffs-Appellants’ Opening Brief (“OB.__”) explained how Defendants
acted with scienter in undertaking their scheme to defraud, detailing deliberate
misconduct that included false-and-misleading statements as well as manipulation of
key metrics to facilitate $2.2 billion of insider selling.
In response, Defendants-Appellees’ Answering Brief (“AB.__”)
mischaracterizes Plaintiffs’ allegations, relies upon inapposite case law, and lauds the
district court’s erroneous pleading demands.
As explained below, none of their gambits prevail.
A. Plaintiffs’ allegations confirm that Defendants acted with
scienter in connection with their scheme to defraud, false-
and-misleading statements, and insider selling.
1. Defendants’ arguments revolve around a specious
theme and several false premises.
Before addressing Defendants’ arguments, it’s helpful to note their underlying
theme, to wit: that Plaintiffs argue that scienter inferences need only be “plausible,”
versus the cogent-and-compelling “strong-inference” standard required by Tellabs,
Inc. v. Makor Issues & Rts., Ltd., 551 U.S. 308 (2007). See AB.53-54 (“[Plaintiffs]
incorrectly cite Bell Atl. Corp. v. Twombly, 550 U.S. 544 (2007) and Ashcroft v. Iqbal,
556 U.S. 662 (2009) several times for the proposition that all Plaintiffs must do is
‘plausibly’ allege scienter.”); accord AB.18; AB.23.
-1-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 8
Defendants’ theme is specious. The Supreme Court and this Court recognize
that allegations’ “plausibility” co-exists with the PSLRA’s higher scienter-pleading
standard. See, e.g., Tellabs, 551 U.S. at 323 (“in determining whether the pleaded
facts give rise to a ‘strong’ inference of scienter, the court must take into account
plausible opposing inferences”);1 Clinton v. Sec. Benefit Life Ins. Co., 63 F.4th 1264,
1281 (10th Cir. 2023) (“particularized allegations plausibly allege [defendant]
engaged in a fraudulent scheme”); accord Ind. Pub. Ret. Sys. v. Pluralsight, Inc., 45
F.4th 1236, 1268 (10th Cir. 2022).2
Defendants build upon their specious theme utilizing three false premises:
(i) Palantir’s present condition and stock price somehow insulate Defendants’ Class
Period misconduct years ago; (ii) an unpublished decision in unrelated state-court
litigation means the district court here correctly dismissed the Complaint; and
(iii) because this appeal centers on scienter, Defendants’ alleged falsehoods don’t
figure into the Court’s analysis.
None of the three premises should sway this Court.
1
Throughout this brief internal citations and footnotes are omitted and emphasis
added, unless noted otherwise.
2
Plaintiffs also cited Iqbal and Twombly for a second axiom: judicial disbelief of
factual allegations is reversible error. OB.40; OB.58; OB.64. Defendants offer no
real defense of that district-court error—nor could they. Twombly, 550 U.S. at 555-
56.
-2-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 9
As to the first, it’s not only common sense but black-letter law that “[f]raud
depends on the state of events when a statement is made, not on what happens later.”
Schleicher v. Wendt, 618 F.3d 679, 684 (7th Cir. 2010). Relatedly, that Palantir stock
now trades higher than at the Class Period’s end isn’t relevant; whatever Palantir
investors did (or didn’t do) with their stock following the fraud’s exposure doesn’t
undercut the fraud’s existence. See, e.g., Harris v. Union Elec. Co., 787 F.2d 355, 372
(8th Cir. 1986) (by holding onto defendant’s securities after fraud’s discovery,
investors made “second investment decision”).
As to the second premise, the state-court decision doesn’t control here. It was
issued in a different venue, in a separate lawsuit brought by different plaintiffs
asserting state-law claims. Cent. Laborers’ Pension Fund v. Karp, 2025 WL 1213104
(Del. Ch. Apr. 25, 2025). Defendants also entirely ignore that the dismissal was on
“demand futility” grounds under Delaware Chancery Court Rule 23.1, and not for
failure to state a claim. Id. at *22. There’s no legal effect upon this litigation.
As to the third, Defendants suggest that because the parties’ briefing focuses on
scienter, Defendants’ alleged misstatements regarding Palantir’s “growth, unit
economics and margins, bespoke versus SaaS-like offerings, and efficiencies” are “not
at issue.” AB.55 n.41.
-3-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 10
Wishful thinking, for a trio of reasons: (i) Plaintiffs fully briefed Defendants’
misstatements in the Opening Brief and explained why they were false-and-
misleading; (ii) the misstatements provide this Court valuable context for ruling upon
the scienter issues (see, e.g., Adams v. Kinder-Morgan, Inc., 340 F.3d 1083, 1106
(10th Cir. 2003)); and (iii) Defendants’ failure to address Plaintiffs’ scienter
allegations in connection with those misstatements constitutes waiver. Clem v.
Lomeli, 566 F.3d 1177, 1182 (9th Cir. 2009) (appellee waives issue by not addressing
it in answering brief).
With Defendants’ specious theme and false premises dispatched, we now turn
to their remaining scienter arguments.
2. Pluralsight confirms Defendants’ billions of dollars in
insider sales at inflated prices bolster a strong
inference of scienter.
Plaintiffs explained how Defendants’ more-than-$2-billion in insider selling
satisfied this Court’s Pluralsight decision, where far-less-dramatic selling—$22
million for one defendant, and just $519,000 for another—helped raise a strong
inference of scienter. See OB.42-44; see also Pluralsight, 45 F.4th at 1264-67
(considering “‘amount of profit from the sales, the portion of stockholdings sold, the
change in volume of insider sales, and the number of insiders selling,’” along with
greater volume of sales inside class period “than outside”). Plaintiffs also explained
-4-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 11
the district court’s myriad errors while considering Defendants’ sales (OB.52-56),
along with its failure to “address, let alone analyze, Pluralsight’s scienter holdings.”
OB.47.3
Defendants respond with a blunderbuss attack wrapped in arguments parroting
the district court’s erroneous analyses. Each scattershot volley misses its mark.
One missed shot concerns ostensibly pre-placed stock-trading plans.
Defendants contend that most of their insider selling was conducted via Rule 10b5-1
“trading plans,” and therefore is immunized. AB.27-29. That untested contention has
multiple flaws.
First, the district court refused to entertain that argument. Add:23 n.10 (“[T]he
Court does not consider Defendants’ automatic trading argument in its scienter
analysis.”). That refusal was for good reason—rather than submit their supposed
trading plans, Defendants proffered “exhibits consisting of SEC filings and attorney-
created documents aggregating data from those filings.” Add:22 n.10. Those filings
were mostly Form 4 trading records (id.)—the truth of which isn’t judicially
3
It’s worth repeating that the district court mentioned this Court’s on-point
Pluralsight opinion only twice. OB.38-39.
-5-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 12
noticeable. Kramer v. Time Warner, Inc., 937 F.2d 767, 774 (2d Cir. 1991); accord
Lovelace v. Software Spectrum, 78 F.3d 1015, 1018 (5th Cir. 1996).4
In addition, this Court holds that Rule 10b5-1 trading plans aren’t a per se
defense to insider-selling charges. Pluralsight, 45 F.4th at 1265-66. As this Court
noted, experience shows such plans are easily manipulated:
As Plaintiffs correctly explain, the text and history of Rule 10b5-1 shows
that such plans can be manipulated easily for personal financial gain
and thus cannot rebut the inference that personal financial gain was a
motive for Defendants’ material misrepresentations.
Id. at 1266. Whether that occurred here is a fact-specific inquiry whose resolution is
impossible given one glaring fault: despite pushing their trading-plan theory,
Defendants never produced any of the plans’ details. As Plaintiffs explained below,
Defendants made no showing as to the plan details, and they: (i) admitted that not all
sales occurred pursuant to a 10b5-1 plan; (ii) failed to provide the district court with
date(s) on which any 10b5-1 plan was adopted or their terms; and (iii) had not
established that they were unaware of their schedule of trades. Appdx.V9.1970.
Given that missing information, it’s no wonder the district court refused to entertain
Defendants’ argument. Add:22 n.10.
4
Defendants say Plaintiffs don’t contest the filings’ validity (AB.27-28), but
Plaintiffs surely did. See ECF113:15 (“Defendants only offer the Forms 3 and 4 for
the improper inferences and arguments that they derive from them—which are not
proper subjects of judicial notice.”).
-6-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 13
One final point regarding Defendants’ trading-plan argument: they’ve veered
dangerously close to misleading this Court by asserting that Pluralsight’s trading-plan
“holding is inapplicable, as the District Court rightly concluded.” AB.29. That
assertion is unsupportable: the district court said it wasn’t considering any trading-
plan argument—and it didn’t distinguish Pluralsight’s “per se” holding or conclude
that it didn’t apply. Add:22 n.10.
Elsewhere, Defendants insist the district court never substituted its “‘benign
view’” of their multi-billion-dollar insider selling for Plaintiffs’ contrary allegations.
AB.26. Yet that’s precisely what happened: the court deemed it “unsurprising” that
Palantir insiders would cash out after taking the Company public while “candidly”
explaining their actions. OB.52-53. With that reasoning the court erred twice.
First, accepting Defendants’ excuse versus the Complaint’s contrary
allegations—inflated-stock sales as part of an overall scheme—is erroneous.
Goldman v. Belden, 754 F.2d 1059, 1070-71 (2d Cir. 1985) (indulging defendant’s
excuses for stock sales as “‘logical’” and “‘reasonable’” “impermissibly reached
beyond” complaint’s scope); see also Helwig v. Vencor, Inc., 251 F.3d 540, 558 (6th
Cir. 2001) (en banc) (rejecting defendant’s self-serving explanation at the pleading
-7-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 14
stage).5 Moreover, the district court’s benign view sweeps too broadly. It gives
corporate insiders carte blanche to defraud when taking a company public, so long as
they announce they’re tapping newfound liquidity. That has never been the law. 6
Nor do Defendants have a compelling defense to the stark contrast between
their $2.2-billion-plus insider selling proceeds and their annual compensation.7
(Recall, those proceeds represented an astronomical 13,900% to 95,200% more than
that compensation. Appdx.V1.129:¶190.) Defendants simply repeat the district
court’s rationale that tech executives are often paid with equity. AB.31.
But that rationale doesn’t acknowledge the reality undermining it here: if equity
compensation was Defendants’ innocent motive, how to explain their actions after the
Class Period, when their insider selling dried up? Appdx.V1.129-30:¶192. That
cessation only adds to the strong-scienter inference. Pluralsight, 45 F.4th at 1267
5
Defendants’ half-hearted response to Helwig and Goldman’s holdings observes
only that neither involved public offerings. AB.27 n.9.
6
The court’s rationale also ignores that Defendants’ inflated-stock sales weren’t
confined to the Offering; instead, Defendants embarked upon a 13-month selling
binge while driving Palantir’s stock price higher with continuing misconduct.
7
Defendants accuse Plaintiffs of conflating “proceeds” with “profits.” AB.25 n.7.
That accusation rings hollow, as the Opening Brief repeatedly speak of the proceeds
Defendants reaped. E.g., OB.14; OB.35; OB.43.
-8-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 15
(suspicious when “volume of sales within the class period was dramatically higher
than outside it”).8
Defendants really miss the mark concerning the district court’s retained-equity
analysis. AB.29. For example, they don’t address the court’s error in reasoning that
“only” Glazer sold “a majority of his holdings” (Add:21)—when this Court doesn’t
require insiders to have sold a “majority” of their holdings to find the sales suspicious.
Pluralsight, 45 F.4th at 1267 (5% and 12% deemed suspicious). Defendants also
don’t address Plaintiffs’ argument that the district court didn’t even mention
Pluralsight while discussing percentages—error again. OB.55.
Compounding those shortcomings, Defendants claim that in calculating their
retained equity, Plaintiffs disregarded “unvested options and [restricted stock units]”
that courts supposedly “routinely consider.” AB.29 (citing Ronconi v. Larkin, 253
F.3d 423, 435-36 (9th Cir. 2001), abrogation recognized by Cai v. Eargo, Inc., 2025
WL 66041 (9th Cir. Jan. 10, 2025), and Smallen v. W. Union Co., 950 F.3d 1297,
1310-11 (10th Cir. 2020)). AB.29.
8
The reduced sales also undermine Defendants’ “trading plan” defense. Wouldn’t
those plans still have been in place despite Palantir’s stock-price movements? Or
were Defendants manipulating the plans, too, while avoiding lower stock prices? The
resulting inference cuts against Defendants.
-9-
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 16
Defendants hoist themselves on their own petard. Unvested securities like
granted options or restricted stock units cannot be exercised or sold until they vest!9
Defendants’ authority even highlights that key distinction. See Ronconi, 253 F.3d at
435 n.25 (“Stock options should be considered in calculating the percentage of shares
sold unless the insider could not have exercised them.”). Smallen doesn’t
acknowledge Ronconi’s distinction, but simply repeats that options are included in
overall holdings. Smallen, 950 F.3d at 1310.10
Defendants have another problem with their retained-equity argument. Even
crediting the figures they list at AB.29—which Plaintiffs don’t—Defendants concede
still-dramatic selling percentages: 23% for Thiel, 19% for both Karp and Cohen, 24%
9
Restricted stock units “have no tangible value until they are vested.” Jason
Fernando, Restricted Stock Unit (RSU): How It Works and Pros and Cons,
Investopedia (Sep. 18, 2025), https://www.investopedia.com/terms/r/restricted-stock-
unit.asp. Similarly, employee stock options “vest over a designated period of time.
Vesting means it becomes available to use.” James Chen, Understanding Stock
Options: Trading Basics and Practical Examples, Investopedia (Aug. 17, 2025),
https://www.investopedia.com/terms/s/stockoption.asp.
10
Defendants inadvertently concede that unvested options shouldn’t be considered
part of their “retained equity.” Despite claiming that Karp “retained” 81% of his
equity (AB.29), Defendants elsewhere argue that the August 2020 stock awards to him
don’t add to the scienter inference because “more than 90% of Karp’s options did not
vest” by the Class Period’s end “and thus could not be sold” until then. AB.32.
- 10 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 17
for Sankar, and 55% for Glazer.11 Defendants’ self-serving percentages still dwarf the
numbers this Court in Pluralsight found suspicious—as does their overall $2.2 billion
in insider selling, which is nearly 100 times the sum that helped raise suspicion in
Pluralsight.
3. Defendants’ construction and utilization of the SPAC
scheme was deliberate misconduct.
Defendants’ primary defense to the round-trip SPAC scheme they concocted
(OB.17-25) centers on disclosure—that they supposedly disclosed all material
information about the deals, and thus didn’t mislead investors. AB.34-36.
That’s untrue. Analysts covering Palantir certainly found the disclosures
lacking. Following Palantir’s May 9, 2022 announcement that the SPAC program had
been wound down in the prior quarter, analysts carped they lacked the ability to
calculate the underlying businesses’ strength. See Appdx.V1.126:¶182. Indeed,
William Blair and RBC analysts complained they were “disappointed with fewer
disclosures in the 10-Q,” and that they had been forced to guess about the state of
some of Palantir’s SPAC investments, without knowing some had already been
terminated. Id. It wasn’t until three months later that RBC published a report that
11
Plaintiffs’ allegations of Defendants’ insider selling (including exercisable options
and vested RSUs) show dramatic percentages: 22.5% (Thiel); 27.8% (Cohen); 40.6%
(Karp); 41.3% (Sankar); and 71.7% (Glazer). Appdx.V1.128-29:¶189. Those
percentages are taken as true. Pluralsight, 45 F.4th at 1250.
- 11 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 18
provided a deep dive into Palantir’s SPAC-related misconduct, and revealed how
poorly the SPACs had fared. Appdx.V1.127:¶185. That savvy analysts found
Palantir’s SPAC-related disclosures deficient speaks volumes. Dirks v. SEC, 463 U.S.
646, 658-59 (1983) (securities analysts “‘ferret out and analyze [corporate]
information’” that “normally may be the basis for judgments as to the market worth of
a corporation’s securities”).
Moreover, while Defendants describe some general SPAC-related information
they supposedly disclosed (AB.34), the Complaint alleges other material facts that
weren’t disclosed. Significantly, Defendants didn’t disclose that because the SPAC
deals carried much-lower costs—90% lower than Palantir’s traditional commercial
deals—Palantir’s reported margins were materially inflated. Appdx.V1.91-95:¶¶103-
109. Other SPAC-related non-disclosures also misled investors. See, e.g.,
Appdx.V1.73:¶63 (although first SPAC deal executed in March 2021, no disclosures
until May 2021—and even then, misleadingly described as a “long-term strategy”);
Appdx.V1.77:¶73 (referencing just four SPAC deals while concealing the massive
SPAC Scheme expansion that was underway); Appdx.V1.79:¶76 n.16 (Defendants
kept two SPACs’s identities under wraps for six months); Appdx.V1.83:¶84 (at least
four SPACs were teetering on insolvency, as some publicly stated soon thereafter).
- 12 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 19
Defendants’ counter-narrative claims that the SPACs “used Palantir’s software”
(AB.35), implying the deals were specifically tailored and working well, but the
Complaint alleges otherwise. Palantir gave the SPACs off-the-rack, partial versions
of the complete Foundry software package, without providing the necessary support to
make Foundry work. Appdx.V1.85-86:¶89; Appdx.V1.91-92:¶103. The Complaint
alleges at least four SPACs that either couldn’t, or didn’t, use Foundry as Defendants
represented. Appdx.V1.75-76:¶69 (Lilium); Appdx.V1.76:¶70 (Sarcos);
Appdx.V1.81:¶80 (Celularity); Appdx.V1.81-82:¶81 (Wejo); see also Appdx.V1.85-
86:¶89 (Celularity’s Foundry-specific complaints). These facts must be accepted as
true. Tellabs, 551 U.S. at 322.
Defendants also insist the quid pro quo SPAC deals were “[l]egitimate,” and
that it’s “undisputed that Palantir received equity and cash consideration from the
SPACs.” AB.35-36. Not so.
First, and most notably, Defendants don’t defend the district court’s error in
requiring that sham deals be “obviously lacking in commercial justification.” Add:23.
That heightened criterion ignored well-pleaded allegations that only needed to
“plausibly” allege the SPAC deals’ fraudulent nature. Iqbal, 556 U.S. at 679; see also
Clinton, 63 F.4th at 1281 (“particularized allegations plausibly allege[d] … a
fraudulent scheme”).
- 13 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 20
Moreover, Palantir’s receipt of cash from the SPACs isn’t a defense; in fact,
receiving cash back after providing the SPACs with funding was the entire point of
the quid pro quo scheme! See, e.g., Appdx.V1.78:¶76 (Boxed agrees to pay back $20
million following Palantir’s $20 million infusion). And the notion that Palantir
received material “equity” in the SPACs is laughable: many of the 20-odd SPACs
were spiraling toward insolvency—seven of them going bankrupt. Appdx.V1.71-
72:¶60; see also Appdx.V1.127:¶185 (“at least 15 SPAC customers … have
substantial doubt as going concerns”).
Trying to combat the damning inference arising out of the SPACs paying
millions of dollars back to Palantir, Defendants say some paybacks were “significantly
different” than Palantir’s cash infusions. AB.36 & n.23. Some were—but sauce for
the goose is sauce for the gander: several others involved dollar-for-dollar round-trips.
See, e.g., Appdx.V1.78:¶76 (Boxed agrees to pay back the $20 million after Palantir’s
$20 million infusion); Appdx.V1.87:¶93 (BlackSky assumes an $8 million Foundry
subscription after Palantir hands it $8 million); Appdx.V1.87:¶94 (Rigetti given $10
million in exchange for what appears to be an estimated $10 million Palantir
subscription agreement). Other deals made no financial sense. See, e.g.,
Appdx.V1.73:¶64 (in exchange for $41 million cash infusion, Lilium agrees to a $50
- 14 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 21
million Foundry subscription—a sum approximately 1,000 times Lilium’s 2021
revenue). 12
Taking aim at allegations that they knew how troubled the SPACs were given
in-depth due diligence and Palantir’s Foundry software, Defendants parrot the district
court’s “‘digital twin’” criticism—i.e., that supposedly only the SPACs, and not
Palantir, had access to the SPACs’s data after Foundry’s installation. AB.37; see
Add:19 n.8 (court rejects alleged access as “pure speculation”).
But Defendants (and the district court) ignore the well-pleaded allegation that,
given Foundry’s “digital twin” function, shortly after each deal closed, Palantir had
as much information about the SPACs as the SPACs themselves. Appdx.V1.71:¶59;
see also Appdx.V1.86:¶90 (Palantir’s Form 10-Q boasts of Company’s “proximity to
these [SPACs]”). Those well-pleaded facts are accepted as true. Pluralsight, 45 F.4th
at 1250.
Concerning “due diligence” allegations, Defendants say the district court simply
followed precedent when it demanded the “‘specific facts or data’” Defendants
gleaned from their pre-deal examinations. AB.37. But that’s untrue; the court
12
Defendants also try to counter the “sham” nature of their deals by noting Plaintiffs
haven’t alleged an accounting restatement or revenue-recognition violation. AB.4;
AB.35 n.21. The suggestion is a straw man, as accounting restatements aren’t
required to show scienter. See, e.g., Aldridge v. A.T. Cross Corp., 284 F.3d 72, 83 (1st
Cir. 2002).
- 15 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 22
demanded far-more-granular, internal data—which isn’t required to sufficiently plead
fraud. Herman & MacLean v. Huddleston, 459 U.S. 375, 390 n.30 (1983)
(“circumstantial evidence” of scienter “can be more than sufficient”); see also
Pirraglia v. Novell, Inc., 339 F.3d 1182, 1193 n.14 (10th Cir. 2003) (demanding
alleged internal paperwork “is asking too much” at the pleading stage). 13
Defendants’ final attack on the SPAC scheme centers on Plaintiffs’ “temporal
proximity” argument (AB.38), which explains that abruptly terminating the SPAC
program shortly after extolling its contributions adds to a strong-scienter inference.
OB.49-50; see, e.g., Helwig, 251 F.3d at 552 (“closeness in time” one of “fixed
constellations of facts … relevant to scienter”). Defendants argue that the six-month
span between their SPAC-program cheerleading and its sudden wind-down is
insufficient as a matter of law (AB.38)—but cite no precedent so holding.
Instead, Defendants lift a couple of time periods from Helwig and two other
decisions—2, 4, and 11 weeks—and present them as the final word on acceptable
temporal proximity. AB.38 n.26. That arbitrary cherry-picking is twice wrong.
13
Defendants’ criticism that Plaintiffs haven’t identified “a single fact” gleaned in
due diligence is unfounded. See, e.g., Appdx.V1.82:¶82 (Glazer participated in
extensive due diligence into Fast Radius, which focused on Fast Radius’s financing
and cash positions for the coming 12 months.). Fast Radius declared bankruptcy a
mere nine months after Palantir’s funding. Id.
- 16 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 23
First, the salient point is that Defendants reassured Palantir investors that the
SPAC deals were part of the Company’s “long-term strategy,” and that “[t]hese are
companies that we think we will be working with for a very long time.”
Appdx.V1.82-83:¶83. Yet within a few months of those reassurances, Palantir shut
down the SPAC program in 1Q22 (Appdx.V1.125-26:¶181)—which shutdown, along
with disappointing revenue-growth figures, helped drive Palantir’s stock down 21%
during a single trading day, with an additional 10% drop over the next two days.
Appdx.V1.126:¶183. The contrast between Defendants’ reassurances and their 180-
degree turnaround is “relevant to scienter.” Helwig, 251 F.3d at 552.
Second, Defendants’ focus on those three time periods is easily countered by
other periods deemed sufficient—illustrating there’s no hard-and-fast temporal-
proximity span as a matter of law. See, e.g., Reese v. Malone, 747 F.3d 557, 574-75
(9th Cir. 2014) (“three to six months” temporal proximity bolsters scienter inference),
overruled in part on other grounds by City of Dearborn Heights Act 345 Police &
Fire Ret. Sys. v. Align Tech., Inc., 856 F.3d 605 (9th Cir. 2017); In re Grand Casinos
Sec. Litig. Inc., 988 F. Supp. 1273, 1283 (D. Minn. 1997) (seven months is “short time
frame” bolstering strong inference of defendants’ knowledge).
Under Tellabs’s holistic approach, temporal proximity remains relevant, and is
yet one more factor bolstering Defendants’ scienter.
- 17 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 24
4. Defendants’ conceded access to truthful information
bolsters a strong inference of scienter.
Defendants have no real answer to Plaintiffs’ charge that the district court erred
when proclaiming that “‘access alone’” wasn’t enough to support a strong inference
of their scienter (Add:19), given that here—unlike in Meitav Dash Provident Funds &
Pension Ltd. v. Spirit AeroSystems Holdings, Inc., 79 F.4th 1209 (10th Cir. 2023)—
Plaintiffs alleged more than mere access. OB.62-64. And Defendants conspicuously
ignore the court’s error in opining that it was “pretty unlikely that Defendants actually
had access” to the data “that Plaintiffs allege they did.” OB.64 (noting erroneous
“judicial disbelief”).
Instead, Defendants assert that Plaintiffs’ “lone response” to the district court’s
access reasoning is that the Complaint attributed certain falsehoods to certain
Defendants—which attribution, they say, has nothing to do with scienter. AB.39.
Defendants’ assertion is incorrect on multiple levels.
First, Defendants themselves repeatedly confirmed their access to relevant
information. See, e.g., Appdx.V1.103:¶126 (Palantir tells investors that its sales
efforts have “historically depended on the significant direct involvement of [its] senior
management team, including Mr. Karp”); see also Appdx.V1.57-58:¶25 (regular
“sales ops” team meetings attended by Karp, Cohen, or Sankar); Appdx.V1.71:¶59
(after installation Foundry unlocks a “digital twin” of each company’s “entire
- 18 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 25
enterprise” and customer data in hours, or even “minutes”);14 Appdx.V1. 102-03:¶125
(on investor/analyst conference call, Defendants emphasized a “number of factors that
provide strong visibility and stability for future revenue growth”); Appdx.V1.103-
04:¶127 (Defendants confirm they’re aware of negative trends in Palantir’s revenues
several periods before publicly reporting them); Appdx.V1.103:¶126 (Karp insisting,
“I know how the business is doing.”).
Second, Defendants’ alleged misstatements did communicate their access to
information contradicting those public statements. See, e.g., Appdx.V1.67:¶47
(“customers are increasingly adopting our software, which can be ready in days”);
Appdx.V1.85:¶88 (Kawasaki insists Palantir is delivering the “full power of Foundry”
to the SPACs); Appdx.V1.107:¶¶135-136 (Sankar boasts that government growth is
“just at the tip of the iceberg,” while commercial segment’s pipeline is ostensibly
“substantial and growing”); Appdx.V1.61-62:¶34 (Sankar claims Palantir now
possesses “SAAS[-]like economics,” and isn’t doing “bespoke applications”
anymore). 15 Public statements like those raise the unassailable inference that
Defendants were privy to data and facts underlying the subjects about which they
14
Plaintiffs dealt with the Defendants’ incorrect “digital twin” argument supra.
15
At the Class Period’s end, Sankar admitted that was false; Palantir had continued
building bespoke applications all along. Appdx.V1.62:¶35.
- 19 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 26
spoke. E.g., Omnicare, Inc. v. Laborers Dist. Council Constr. Indus. Pension Fund,
575 U.S. 175, 189 (2015) (investors reasonably believe issuer’s statement “fairly
aligns with the information in the issuer’s possession at the time”). 16
Third, while Defendants dispute or deny Plaintiffs’ well-pleaded factual
allegations going to their access, it’s axiomatic that those allegations are accepted as
true, with the resulting inferences taken in Plaintiffs’ favor. Tellabs, 551 U.S. at 322.
In refusing to do so, the district court committed reversible error. Pluralsight, 45
F.4th at 1261, 1263.
Trying to chip away at “‘access,’” Defendants paint as commonplace corporate-
speak CEO Karp’s insistence that “‘I run this business, for better or worse. I know
how the business is doing.’” AB.41.
Karp’s admission of his involvement and knowledge is accepted at face value,
and goes directly to his scienter. See, e.g., Pluralsight, 45 F.4th at 1264 (defendant
admitted monitoring key metric that he “regularly reported to investors and analysts”);
16
Defendants say Omnicare is “no help” because it was addressing non-fraud claims
and falsity under Securities Act §11. AB.40 n.27. They’re mistaken; that investors
understand corporate public statements to reflect “the information in the [speaker’s]
possession at the time” (Omnicare, 575 U.S. at 189) is surely a universal truth across
all types of securities lawsuits involving both fraud and non-fraud claims. See, e.g.,
City of Warren Police & Fire Ret. Sys. v. Prudential Fin., Inc., 70 F.4th 668, 685 (3d
Cir. 2023) (“every other Court of Appeals to encounter the issue has applied the
Omnicare framework for opinion falsity to claims for Rule 10b-5 violations”).
- 20 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 27
see also In re Daou Sys., 411 F.3d 1006, 1022 (9th Cir. 2005) (“specific admissions
from top executives that they are involved in every detail of the company … are
factors in favor of inferring scienter”); Lormand v. US Unwired, Inc., 565 F.3d 228,
254 (5th Cir. 2009) (defendants’ admissions “directly and cogently tend to prove”
their state of mind).17
Defendants also try to soft pedal their admission that Palantir’s sales efforts
depended upon the “significant direct involvement” of the Company’s “‘senior
management team’” while singling out CEO Karp. AB.41; see Appdx.V1.103:¶126.
Defendants say the admission is much ado about nothing because they simultaneously
noted that the direct involvement had been done historically—implying the
executives’ participation was long past. AB.41.
It takes more than a self-serving implication to undercut the Complaint’s
allegations. Employees’ Ret. Sys. of Gov’t of the Virgin Is. v. Blanford, 794 F.3d 297,
307 (2d Cir. 2015) (defendants’ excuses “[are] entitled to little weight at this stage of
litigation, when we must ‘accept all factual claims in the complaint as true and draw
all reasonable inferences in the plaintiff’s favor’”). Whatever was happening with
17
Under this Court’s precedents, that Karp was Palantir’s CEO also adds to the
strong-scienter inference. Adams, 340 F.3d at 1106 (“that [defendant] was the most
senior executive of the Company is a fact relevant in our weighing of the totality of
the [scienter] allegations”).
- 21 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 28
Palantir’s sales-force ranks during the Class Period, 18 the logical inference arising out
of Defendants’ assurances about senior management’s involvement was that top
executives had been, and were continuing to be, “significant[ly]” and “direct[ly]”
involved in Palantir’s sales efforts. Appdx.V1.103:¶126. The opposing inference—
that their “significant direct involvement” suddenly halted the day Palantir went
public—lacks any record support and is nonsensical. In fact, Palantir repeated this
representation in every 10-Q and 10-K filed during the Class Period.19
Finally, Defendants spend several pages dissecting the Meitav decision (AB.42-
44), but scrupulously avoid specific language that the district court deemed so
important that it block-quoted the text:
[I]t would make little sense to draw a strong inference of scienter from
access to information. If access alone were enough, a strong inference of
scienter would exist for high-level executives whenever they make a
public statement contradicting something in the company’s files.
Add:19 (quoting Meitav, 79 F.4th at 1217).
18
Plaintiffs do not “concede” that Palantir was successfully doubling its sales force
during the Class Period. AB.41. The Complaint only notes Karp’s claim in February
2021 that Palantir “was in the midst of an effort” to do so. Appdx.V1.65:¶40 n.10.
19
A quick search reveals that Palantir continued to make this representation in its 10-
Ks and 10-Qs long after the Class Period ended, including in its most-recent 10-Q,
filed with the SEC on August 5, 2025. See Palantir Technologies Inc., Quarterly
Report (Form 10-Q) (Aug. 5, 2025)
https://www.sec.gov/ix?doc=/Archives/edgar/data/0001321655/000132165525000106
/pltr-20250630.htm, at 47.
- 22 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 29
Defendants’ avoidance is understandable, perhaps, because Plaintiffs have
alleged much more than access-alone facts: e.g., Defendants’ self-proclaimed “strong
visibility” and “significant direct involvement” of senior executives, Karp’s own
words, Palantir’s admitted proximity to the SPACs, etc. Plaintiffs then complemented
those facts with additional allegations of Defendants’ deliberate actions: e.g.,
breathtaking insider selling of $2.2 billion-plus, scheming conduct vis-à-vis the round-
trip SPAC transactions, manipulation of reported metrics, etc.—all of which are
(i) taken as true, and (ii) considered collectively. Tellabs, 551 U.S. at 322-23;
Pluralsight, 45 F.4th at 1259, 1261.
When they are, Meitav’s fact-specific “access alone” criticism has no
application here.
5. Defendants’ trumpeted “strong visibility” into
Palantir’s revenue growth further establishes their
scienter.
Defendants claim Plaintiffs “do not dispute” the district court’s holding that
“‘strong visibility’” isn’t enough to find that they misled Palantir investors (AB.45),
but nothing is further from the truth. Time and again Plaintiffs argued that
Defendants’ constant claims of “strong visibility” into growth and revenue provided
direct on-ramps to their knowledge that the two areas were not progressing as
represented. See, e.g., OB.26 (the opposite of Defendants’ representations was true,
- 23 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 30
and Defendants were aware of this at the time they spoke, given their “strong
visibility” into Palantir’s future revenue); accord OB.40; OB.41; OB.46; OB.50;
OB.61.
Indeed, Plaintiffs reject the district court’s reasoning that Defendants’ admitted
“strong visibility” cannot satisfy the “total visibility” it demanded. OB.60 (original
emphasis). The court’s demand is doubly wrong: (i) it undermines the “strongly
support[ed]” scienter inferences arising from Defendants’ admissions (Pluralsight, 45
F.4th at 1259-60); and (ii) improperly elevates the requisite inference to an
“irrefutable” level. Tellabs, 551 U.S. at 324. 20
Defendants say Plaintiffs are asking this Court to infer that they had an
“omniscient view” of Palantir’s financial results in advance (AB.46), but that’s untrue.
Plaintiffs aren’t urging omniscience, but advance the commonsense argument that
Defendants’ self-professed “strong visibility”—specifically based on the nature of
their business and contracts—meant they saw Palantir’s existing revenue, growth, and
SPAC-related problems. See, e.g., Appdx.V1.103-04:¶127 (Defendants explain that
20
Defendants defend the district court’s “‘total visibility’” demand as merely
“semantic.” AB.45 n.33. If Defendants mean the two levels of visibility are
fundamentally equivalent while using different words, they’re mistaken: “strong” isn’t
“total” by anyone’s measure.
- 24 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 31
revenue-related problems arising in the present quarter won’t be “fully reflected in
our results of operations until future periods”).
Defendants wind up their “‘total visibility’” screed by charging Plaintiffs with
ignoring Circuit precedent involving “growth forecasts.” AB.45, 46 & n.35. But
Plaintiffs’ case doesn’t concern Defendants’ forecasts, but rather their real-time
knowledge of, and misstatements concerning: (i) stagnating growth; (ii) revenue
impact; and (iii) the sham SPAC deals Defendants used to prop up (i) and (ii).
Defendants would have this Court to leave common sense at the courthouse steps, yet
Palantir’s SPAC customers’ myriad bankruptcies and insolvencies didn’t happen
overnight. Pluralsight, 45 F.4th at 1259-60; see also Plotkin v. IP Axess Inc., 407
F.3d 690, 698 (5th Cir. 2005) (“not unwarranted to infer that when a company’s big
deal collapses so fast, something was amiss at the outset”).21
6. Defendants’ deliberate manipulation of key reporting
metrics cannot be explained away.
Plaintiffs’ Opening Brief tabulates the five key metrics whose reporting
Defendants manipulated to further their scheme and conceal stagnation in Palantir’s
21
Defendants say Plaintiffs overlooked new contracts that Palantir “hoped to sign” or
existing contracts that “could be renewed or upsized.” AB.47. Their charge is
toothless—“hope[]” and “could be” may be understandable aspirations, but cannot
excuse Palantir’s existing reality of stagnating growth, quid pro quo revenue from
unsustainable contracts, and manipulated metrics used in Palantir’s public
(mis)statements.
- 25 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 32
two main business segments. OB.29-33. The brief further explains the district court’s
erroneous reasoning that because some truthful information was available elsewhere in
Palantir’s filings or public statements, their manipulative conduct was excused.
OB.64-66; see also Va. Bankshares, Inc. v. Sandberg, 501 U.S. 1083, 1097 (1991) (“If
it would take a financial analyst to spot the tension between the one and the other,
whatever is misleading will remain materially so, and liability should follow.”).
Responding, Defendants criticize Plaintiffs for not offering any “[confidential
witnesses], internal document, or any facts—as to why these changes were
implemented.” AB.48 (original emphasis). Their criticism, untethered to precedent,
is fanciful and misplaced.
Plaintiffs did explain why Defendants made the deliberate changes. See OB.29-
33 (detailing manipulated performance metrics, and explaining how each concealed
stagnation in Palantir’s business). And neither confidential witnesses nor internal
documents are required to plead a strong inference of scienter. See, e.g., In re
Alphabet, Inc. Sec. Litig., 1 F.4th 687, 707 (9th Cir. 2021) (“confidential witnesses are
not needed where, as here, other allegations in the complaint raise a strong inference
of scienter”); see also Pirraglia, 339 F.3d at 1193 n.14 (internal documents not
needed at the pleading stage).
- 26 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 33
Undaunted, Defendants proceed to give their own narrative of why each of the
reporting metrics was suddenly changed. Their excuses are not only premature, but
unpersuasive.
As to Defendants’ sudden concealment of the “Acquire, Expand, and Scale”
information that undergirded Palantir’s entire business model (Appdx.V1.118-
20:¶¶164-168), Defendants claim that continuing to disclose those categories wouldn’t
have revealed anything “about non-SPAC clients” because SPACs would’ve still been
represented in the metric. AB.49. Even if true, that doesn’t mean that Palantir
investors wouldn’t realize that non-SPAC customers were also languishing in the first
or second phase; logically, investors would’ve understood precisely what the cessation
obscured: Palantir’s diminishing business prospects for its organic (i.e., non-SPAC)
customers, including the: (i) revenue and contribution margins associated with each
cohort of new customers acquired in FY21; (ii) rate at which Palantir’s customers
signed in 2021 had failed to advance through the cohorts to profitability; and
(iii) SPAC Scheme’s effect in masking this downward trend. Appdx.V1.120:¶168. In
other words, Palantir’s declining performance would have been obvious.
As to the newly-expanded definition of “customer,” Defendants say they
provided “more[-]meaningful” period-over-period comparisons, and that Plaintiffs
don’t challenge that explanation. AB.49. But Plaintiffs certainly do—alleging that
- 27 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 34
the change was deliberately done to: (i) artificially inflate Palantir’s existing customer
count; and (ii) rope in revenues from the temporarily boosted financial performance in
2Q20, 3Q20, and 4Q20 that reflected short-term COVID-19-related contracts.
Appdx.V1.117:¶162.
As to obscuring the Total Remaining Deal Value breakdown between
government and commercial sectors (OB.32), Defendants say TRDV was disclosed on
nearly all earnings calls during the Class Period, and again at year-end. AB.50. But
that doesn’t address, let alone excuse, the fact that Defendants deliberately hid the
important metric for three straight quarters. Appdx.V1.120-21:¶170. Whether the
lengthy concealment mattered to investors, or was cured by year-end reports, cannot
be decided now. Pirraglia, 339 F.3d at 1192 n.13.22
As to changing “average revenue per customer” from all customers to just
Palantir’s top 20 clients after the former figure dwindled (OB.32-33), Defendants say
they disclosed the change’s true impetus: acquiring more customers (especially
growth-stage ones) supposedly meant that only the top 20 demonstrated their
22
Defendants offer the same excuse for omitting average contract duration despite
reliably citing it quarter after quarter as a revenue-visibility metric. Appdx.V1.117-
18:¶163. See AB.50 (Palantir included metric in 2021 annual report). Whether that
single disclosure cured Defendants’ omissions over three straight quarters is not
properly resolved now. Pirraglia, 339 F.3d at 1192 n.13.
- 28 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 35
“‘expanding relationship with existing customers’”—and that Plaintiffs haven’t
contested that explanation. AB.50 (citing Appdx.V1.121-22:¶171).
Defendants again twist the Complaint’s allegations. While Plaintiffs repeat
Palantir’s claim that growing customer counts meant lower average revenues,
Plaintiffs further allege that when revenues dropped, Defendants deliberately shifted
to just the top 20—along with the manipulated Acquire-Expand-Scale reporting—in
order to conceal from investors that Palantir’s newest customers were not scaling up
in size or profitability. Appdx.V1.121-22:¶171. That allegation, accepted as true, is
at least as compelling as Defendants’ self-serving excuse. Tellabs, 551 U.S. at 324;
Pluralsight, 45 F.4th at 1268.23
B. Defendants haven’t satisfactorily defended the district
court’s failure to address the deliberate actions comprising
their scheme.
While Plaintiffs explained the district court’s erroneous scienter analyses
(OB.52-68), they also described what the court didn’t do—namely, that it failed to
23
Defendants suggest that investors could have made “straightforward calculation[s]”
to discover what the actual average-revenue-per-customer was (AB.51), but cite no
authority requiring investors to dig through complex data to double-check corporate
representations. Cf. In re Genius Brands Int’l, Inc. Sec. Litig., 97 F.4th 1171, 1186
(9th Cir. 2024) (“securities issuers should not escape liability for misrepresentations
merely because they can show that corrective information was publicly available on
some webpage tucked in a deep corner of the internet or buried in some unwieldy
spreadsheet”). Moreover, even if Palantir investors had attempted that excavation, the
truth would have been thwarted by the Company’s ever-changing definition of
“customer.”
- 29 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 36
address how the deliberate actions comprising Defendants’ scheme and wrongful
course of conduct couldn’t be excused as mere negligence, and thus were intertwined
with their overall scienter. OB.48-52. Under this Court’s precedents, that failure was
error. See Clinton, 63 F.4th at 1288-89 (reversing district court that “ignored other
allegations” relevant to scienter associated with fraudulent scheme and
misstatements); Johnson v. Reyna, 57 F.4th 769, 776 (10th Cir. 2023) (court
“improperly construed Johnson’s complaint by overlooking his plain-language
allegations”).
Defendants grudgingly admit that the district court “did not reach Plaintiffs’
falsity or scheme allegations.” AB.55. Their follow-on insistence that the court
wasn’t “required to reach” the alleged “misstatements or conduct” to analyze their
scienter (id.) is dead wrong. See, e.g., Tellabs, 551 U.S. 326 (court considering
scienter allegations must “assess all [of] the allegations”).
Defendants also assert that even “‘deliberate’” conduct isn’t necessarily
fraudulent. AB.54. That may be so in a different context, but their actions here are
alleged to be fraudulent—and, when considered together “collectively” and
“holistically” (Tellabs, 551 U.S. at 326), combine to form a strong inference of
Defendants’ scienter. Plotting for massive financial gain via quid pro quo SPAC deals
while concealing stagnated growth using manipulated metric-reporting levers doesn’t
- 30 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 37
happen innocently, or even recklessly. Defendants’ deliberate misconduct was
paradigmatic securities fraud.
Defendants also insist that the district court did do the comparative analysis that
Tellabs mandates (AB.54), but that’s overt question-begging. In truth, the court’s
abbreviated analysis was infected by (i) improperly-heightened pleading burdens, and
(ii) its disbelief of allegations that it should have accepted as true. Each error warrants
reversal. Tellabs, 551 U.S. at 324 (scienter inferences “need not be irrefutable”);
Twombly, 550 U.S. at 555-56 (judicial disbelief).
Relating to Defendants’ comparative-analysis argument, they assert that the
scienter-ties-go-to-plaintiffs truism is Plaintiffs’ “last-resort position” with no support.
AB.53 n.39. Defendants’ assertion is baffling, as both the truism’s existence—and
this Court’s embrace of it—are undeniable. Pluralsight, 45 F.4th at 1268 (plaintiffs’
scienter inference need only be “‘at least as compelling as any opposing inference one
could draw from the facts alleged’”) (quoting Tellabs, 551 U.S. at 324); see also
Frank v. Dana Corp., 547 F.3d 564, 571 (6th Cir. 2008) (“‘Tellabs now awards the
draw to the plaintiff.’”).
- 31 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 38
II. CONCLUSION
For the reasons given supra and in the Opening Brief, this Court should reverse
the district court’s dismissal.
DATED: October 21, 2025 Respectfully submitted,
ROBBINS GELLER RUDMAN
& DOWD LLP
DARREN J. ROBBINS
JOSEPH D. DALEY
LUKE O. BROOKS
J. MARCO JANOSKI GRAY
ERIKA L. OLIVER
s/Joseph D. Daley
JOSEPH D. DALEY
655 West Broadway, Suite 1900
San Diego, CA 92101
Telephone: 619/231-1058
619/231-7423 (fax)
darrenr@rgrdlaw.com
joed@rgrdlaw.com
lukeb@rgrdlaw.com
mjanoski@rgrdlaw.com
eoliver@rgrdlaw.com
Counsel for Plaintiffs/Appellants
- 32 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 39
CERTIFICATE OF COMPLIANCE WITH TYPE-VOLUME LIMIT
Certificate of Compliance With Type-Volume Limit,
Typeface Requirements, and Type Style Requirements
1. This document complies with the type-volume limitation of Fed. R. App. P.
32(a)(7)(B)(i) because, excluding the parts of the document exempted by Fed. R.
App. P. 32(f) and 10th Cir. R. 32(B):
[X] this document contains 6,482 words, or
[] this brief uses a monospaced typeface and contains <state the number
of> lines of text.
2. This document complies with the typeface requirements of Fed. R. App. P.
32(a)(5) and the type style requirements of Fed. R. App. P. 32(a)(6) because:
[X] this document has been prepared in a proportionally spaced typeface
using Microsoft Word 2016 in 14 point and Times New Roman typeface, or
[] this document has been prepared in a monospaced typeface using
<state name and version of word processing program> with <state number
of characters per inch and name of type style>.
Date: October 21, 2025
s/Joseph D. Daley
JOSEPH D. DALEY
- 33 -
4896-9760-9332.v1
Appellate Case: 25-1178 Document: 32 Date Filed: 10/21/2025 Page: 40
DECLARATION OF SERVICE
I, the undersigned, declare:
1. That declarant is and was, at all times herein mentioned, a citizen of the
United States and employed in the City and County of San Diego, over the age of
18 years, and not a party to or interested party in the within action; that declarant’s
business address is 655 West Broadway, Suite 1900, San Diego, California 92101.
2. I hereby certify that on October 21, 2025, I electronically filed the
foregoing document: PLAINTIFFS-APPELLANTS’ REPLY BRIEF with the Clerk of
the Court for the United States Court of Appeals for the Tenth Circuit by using the
appellate CM/ECF system.
3. I certify that all participants in the case are registered CM/ECF users and
that service will be accomplished by the appellate CM/ECF system.
I declare under penalty of perjury that the foregoing is true and correct.
Executed on October 21, 2025, at San Diego, California.
s/Joseph D. Daley
JOSEPH D. DALEY
- 34 -
4896-9760-9332.v1