ORDER
The opinion filed on August 28, 2003, 341 F.3d 978, shall be amended as follows.
Replace footnote 2, 341 F.3d at 984, with:
Prosser and Keeton say that a plaintiffs consent is probably invalid if the defendant “ought to have known in the exercise of reasonable care” about the mistake. Prosser & Keeton § 18, at 119. Because the Stored Communications Act defines a criminal offense and includes an explicit mens rea requirement, see 18 U.S.C. § 2701(a)(1), we do not think a defendant can be charged with constructive knowledge on a showing of mere negligence. Rather, the defendant must have consciously procured consent through improper means. In this case, the magistrate found that defendants had acted in bad faith. That is enough to charge them with knowledge of Net-Gate’s mistake. See Black’s Law Dictionary 139 (6th ed.1990) (defining “bad faith” as “not simply bad judgment or negligence, but ... conscious doing of a wrong because of dishonest purpose or moral obliquity”).
Replace the fourth full paragraph, 341 F.3d at 985, with:
The United States, as amicus curiae, disputes our interpretation. It first argues that, because subsection (B) refers to “any storage of such communication,” it applies only to backup copies of messages that are themselves in temporary, intermediate storage under subsection (A). The text of the statute, however, does not support this reading. Subsection (A) identifies a type of communication (“a wire or electronic communication”) and a type of storage (“temporary, intermediate storage ... incidental to the electronic transmission thereof’). The phrase “such communication” in subsection (B) does not, as a matter of grammar, reference attributes of the type of storage defined in subsection (A). The government’s argument would be correct if subsection (B) referred to “a communication in such storage,” or if subsection (A) referred to a communication in temporary, intermediate storage rather than temporary, intermediate storage of a communication. However, as the statute is written, “such communication” is nothing more than shorthand for “a wire or electronic communication.”
The government’s contrary interpretation suffers from the same flaw as Fraser’s: It drains subsection (B) of independent content because virtually any backup of a subsection (A) message will itself qualify as a message in temporary, *1070intermediate storage. The government counters that the statute requires only that the underlying message be temporary, not the backup. But the lifespan of a backup is necessarily tied to that of the underlying message. Where the underlying message has expired in the normal course, any copy is no longer performing any backup function. An ISP that kept permanent copies of temporary messages could not fairly be described as “backing up” those messages.
The United States also argues that we upset the structure of the Act by defining “electronic storage” so broadly as to be superfluous and by rendering irrelevant certain other provisions dealing with remote computing services. The first claim relies on the argument that any copy of a message necessarily serves as a backup to the user, the service or both. But the mere fact that a copy could serve as a backup does not mean it is stored for that purpose. We see many instances where an ISP could hold messages not in electronic storage' — for example, e-mail sent to or from the ISP’s staff, or messages a user has flagged for deletion from the server. In both cases, the messages are not in temporary, intermediate storage, nor are they kept for any backup purpose.
Our interpretation also does not render irrelevant the more liberal access standards governing messages stored by remote computing services. See 18 U.S.C. §§ 2702(a)(2), 2703(b). The government’s premise is that a message stored by a remote computing service “solely for the purpose of providing storage or computer processing services to [the] subscriber,” id. §§ 2702(a)(2)(B), 2703(b)(2)(B), would also necessarily be stored for purposes of backup protection under section 2510(17)(B), and thus would be subject to the more stringent rules governing electronic storage. But not all remote computing services are also electronic communications services and, as to those that are not, section 2510(17)(B) is by its own terms inapplicable. The government notes that remote computing services and electronic communications services are “often the same entities,” but “often” is not good enough to make the government’s point. Even as to remote computing services that are also electronic communications services, not all storage covered by sections 2702(a)(2)(B) and 2703(b)(2)(B) is also covered by section 2510(17)(B). A remote computing service might be the only place a user stores his messages; in that case, the messages are not stored for backup purposes.
Finally, the government invokes legislative history. It cites a passage from a 1986 report indicating that a committee intended that messages stored by a remote computing service would “continue to be covered by section 2702(a)(2)” if left on the server after user access. H.R.Rep. No. 647, 99th Cong., at 65 (1986). The cited discussion addresses provisions relating to remote computing services. We do not read it to address whether the electronic storage provisions also apply. See id. at 64-65. The committee’s statement that section 2702(a)(2) would “continue” to cover email upon access supports our reading. If section 2702(a)(2) applies to e-mail even before access, the committee could not have been identifying an exclusive source of protection, since even the government concedes that unopened e-mail is protected by the electronic storage provisions.
The government also points to a subsequent, rejected amendment that would have made explicit the electronic storage definition’s coverage of opened e-mail. See H.R.Rep. No. 932, 106th Cong., at 7 (2000). This sort of legislative history has very little probative value; Con*1071gress might have rejected the amendment precisely because it thought the definition already applied.
Finally, the government points to a passing reference in another recent report. H.R.Rep. No. 236(1), 107th Cong., at 57 (2001). The discussion addresses an unrelated issue and merely assumes that the definition only applies to unopened e-mail. This assumption was natural given that Fraser had recently been decided.
We acknowledge that our interpretation of the Act differs from the government’s and do not lightly conclude that the government’s reading is erroneous. Nonetheless, for the reasons above, we think that prior access is irrelevant to whether the messages at issue were in electronic storage. Because plaintiffs email messages were in electronic storage regardless of whether they had been previously delivered, the district court’s decision cannot be affirmed on this alternative ground.4
The petition for rehearing and rehearing en banc is otherwise DENIED. See Fed. R.App. P. 35; Fed. R.App. P. 40. No further petitions will be accepted.
OPINION
KOZINSKI, Circuit Judge:We consider whether defendants violated federal electronic privacy and computer fraud statutes when they used a “patently unlawful” subpoena to gain access to email stored by plaintiffs’ Internet service provider.
Background
Plaintiffs Wolf and Buckingham, officers of Integrated Capital Associates, Inc. (ICA), are embroiled in commercial litigation in New York against defendant Far-ey-Jones. In the course of discovery, Far-ey-Jones sought access to ICA’s e-mail. He told his lawyer Iryna Kwasny to subpoena ICA’s ISP, NetGate.
Under the Federal Rules, Kwasny was supposed to “take reasonable steps to avoid imposing undue burden or expense” on NetGate. Fed.R.Civ.P. 45(c)(1). One might have thought, then, that the subpoena would request only e-mail related to the subject matter of the litigation, or maybe messages sent during some relevant time period, or at the very least those sent to or from employees in some way connected to the litigation. But Kwasny ordered production of “[a]ll copies of e-mails sent or received by anyone” at ICA, with no limitation as to time or scope.
NetGate, which apparently was not represented by counsel, explained that the amount of e-mail covered by the subpoena was substantial. But defendants did not relent. NetGate then took what might be described as the “Baskin-Robbins” approach to subpoena compliance and offered defendants a “free sample” consisting of 339 messages. It posted copies of the messages to a NetGate website where, without notifying opposing counsel, Kwas-ny and Farey-Jones read them. Most were unrelated to the litigation, and many were privileged or personal.
When Wolf and Buckingham found out what had happened, they asked the court to quash the subpoena and award sanctions. Magistrate Judge Wayne Brazil soundly roasted Farey-Jones and Kwasny for their conduct, finding that “the subpoe*1072na, on its face, was massively overbroad” and “patently unlawful,” that it “transparently and egregiously” violated the Federal Rules, and that defendants “acted in bad faith” and showed “at least gross negligence in the crafting of the subpoena.” He granted the motion to quash and socked defendants with over $9000 in sanctions to cover Wolf and Buckingham’s legal fees. Defendants did not appeal that award.
Wolf, Buckingham and other ICA employees whose e-mail was included in the sample also filed this civil suit against Farey-Jones and Kwasny. They claim defendants violated the Stored Communications Act, 18 U.S.C. § 2701 et seq., the Wiretap Act, 18 U.S.C. § 2511 et seq., and the Computer Fraud and Abuse Act, 18 U.S.C. § 1030, as well as various state laws. The district court held that none of the federal statutes applied, and dismissed the claims without leave to amend. It declined jurisdiction over the state law claims under 28 U.S.C. § 1367(c)(3). Plaintiffs now appeal.
Analysis
1. The Stored Communications Act provides a cause of action against anyone who “intentionally accesses without authorization a facility through which an electronic communication service is provided ... and thereby obtains, alters, or prevents authorized access to a wire or electronic communication while it is in electronic storage.” 18 U.S.C. §§ 2701(a)(1), 2707(a). “[Electronic storage” means either “temporary, intermediate storage ... incidental to ... electronic transmission,” or “storage ... for purposes of backup protection.” Id. § 2510(17). The Act exempts, inter alia, conduct “authorized ... by the person or entity providing a wire or electronic communications service,” id. § 2701(c)(1), or “by a user of that service with respect to a communication of or intended for that user,” id. § 2701(c)(2).
The district court dismissed on the ground that NetGate had authorized defendants’ access. It held that this consent was not coerced, because the subpoena itself informed NetGate of its right to object. Plaintiffs contend that NetGate’s authorization was nonetheless invalid because the subpoena was patently unlawful. Their claim turns on the meaning of the word “authorized” in section 2701. We have previously reserved judgment on this question, see Konop v. Hawaiian Airlines, Inc., 302 F.3d 868, 879 n. 8 (9th Cir.2002), while other circuits have considered related issues, see, e.g., EF Cultural Travel BV v. Explorica, Inc., 274 F.3d 577, 582 n. 10 (1st Cir.2001) (holding access might be “unauthorized” under the Computer Fraud and Abuse Act if it is “not in line with the reasonable expectations” of the party granting permission (internal quotation marks omitted)); United States v. Morris, 928 F.2d 504, 510 (2d Cir.1991) (holding access unauthorized where it is not “in any way related to[the system’s] intended function”).
We interpret federal statutes in light of the common law. See Beck v. Prupis, 529 U.S. 494, 500-01, 120 S.Ct. 1608, 146 L.Ed.2d 561 (2000). Especially relevant here is the common law of trespass. Like the tort of trespass, the Stored Communications Act protects individuals’ privacy and proprietary interests. The Act reflects Congress’s judgment that users have a legitimate interest in the confidentiality of communications in electronic storage at a communications facility. Just as trespass protects those who rent space from a commercial storage facility to hold sensitive documents, cf. Prosser and Keeton on the Law of Torts § 13, at 78 (W. Page Keeton ed., 5th ed.1984), the Act protects users whose electronic communi*1073cations are in electronic storage with an ISP or other electronic communications facility.
A defendant is not liable for trespass if the plaintiff authorized his entry. See Prosser & Keeton § 13, at 70. But “an overt manifestation of assent or willingness would not be effective ... if the defendant knew, or probably if he ought to have known in the exercise of reasonable care, that the plaintiff was mistaken as to the nature and quality of the invasion intended.” Id. § 18, at 119; cf. Restatement (Second) of Torts §§ 173, 892B(2). Thus, the busybody who gets permission to come inside by posing as a meter reader is a trespasser. J.H. Desnick, M.D., Eye Servs., Ltd. v. ABC, 44 F.3d 1345, 1352 (7th Cir.1995). So too is the police officer who, invited into a home, conceals a recording device for the media. Cf. Berger v. Hanlon, 129 F.3d 505, 516-17 (9th Cir.1997), vacated, 526 U.S. 808, 119 S.Ct. 1706, 143 L.Ed.2d 978 (1999), reinstated in relevant part, 188 F.3d 1155, 1157 (9th Cir.1999).
Not all deceit vitiates consent. “[T]he mistake must extend to the essential character of the act itself, which is to say that which makes it harmful or offensive, rather than to some collateral matter which merely operates as an inducement.” Prosser & Keeton § 18, at 120 (footnote omitted). In other words, it must be a “substantial mistake[ ] ... concerning the nature of the invasion or the extent of the harm.” Restatement (Second) of Torts § 892B(2) cmt. g. Unlike the phony meter reader, the restaurant critic who poses as an ordinary customer is not liable for trespass, Desnick, 44 F.3d at 1351; nor, unlike the wired cop, is the invitee who conceals only an intent to repeat what he hears, cf. Dietemann v. Time, Inc., 449 F.2d 245, 249 (9th Cir.1971) (invasion of privacy claim). These results hold even if admission would have been refused had all the facts been known.
These are fine and sometimes incoherent distinctions. See Med. Lab. Mgmt. Consultants v. ABC, 30 F.Supp.2d 1182, 1201-04 (D.Ariz.1998), aff'd, 306 F.3d 806 (9th Cir.2002). But the theory is that some invited mistakes go to the essential nature of the invasion while others are merely collateral. Classification depends on the extent to which the intrusion trenches on “the specific interests that the tort of trespass seeks to protect.” Desnick, 44 F.3d at 1352; see also Lewis v. United States, 385 U.S. 206, 211, 87 S.Ct. 424, 17 L.Ed.2d 312 (1966); Food Lion, Inc. v. Capital Cities/ABC, Inc., 194 F.3d 505, 517-18 (4th Cir.1999).
We construe section 2701 in light of these doctrines. Permission to access a stored communication does not constitute valid authorization if it would not defeat a trespass claim in analogous circumstances. Section 2701(c)(1) therefore provides no refuge for a defendant who procures consent by exploiting a known mistake that relates to the essential nature of his access. ,.
' Under this standard, plaintiffs have alleged facts that vitiate NetGate’s consent. NetGate disclosed the sample in response to defendants’ purported subpoena. Unbeknownst to NetGate, that subpoena was invalid. This mistake went to the essential nature of the invasion of privacy. The subpoena’s falsity transformed the access from a bona fide state-sanctioned inspection into private snooping. See Restatement (Second) of Torts § 174 (addressing “consent induced by fraud or mistake as to the validity of purported legal authority”); cf. Bumper v. North Carolina, 391 U.S. 543, 549, 88 S.Ct. 1788, 20 L.Ed.2d 797 (1968) (“A search conducted in reliance upon a warrant cannot later be justified on the basis of consent if it *1074turns out that the warrant was invalid.”). The false subpoena caused disclosure of documents that otherwise would have remained private; it effected an “invasion ... of the specific interests that the [statute] seeks to protect.” Desnick, 44 F.3d at 1352.
Defendants had at least constructive knowledge of the subpoena’s invalidity. It was not merely technically deficient, nor a borderline case over which reasonable legal minds might disagree. It “transparently and egregiously” violated the Federal Rules, and defendants acted in bad faith and with gross negligence in drafting and deploying it.1 They are charged with knowledge of its invalidity. See Prosser & Keeton § 18, at 119 (consent likely vitiated where defendants “ought to have known in the exercise of reasonable care” of the mistake).2
That NetGate could have objected is immaterial. The subpoena may not have been coercive, but it was deceptive, and that is an independent ground for invalidating consent. See Restatement (Second) of Torts § 892B(2)-(3). It was a piece of paper masquerading as legal process. NetGate produced the sample in response and doubtless would not have done so had it known the subpoena was void — particularly in light of its own legal obligation not to disclose such messages to third parties, see 18 U.S.C. § 2702(a)(1). That NetGate could have objected proves disclosure was not an inevitable consequence, but it was still a foreseeable one (and the intended one).
Allowing consent procured by known mistake to serve as a defense would seriously impair the statute’s operation. A hacker could use someone else’s password to break into a mail server and then claim the server “authorized” his access. Congress surely did not intend to exempt such intrusions — indeed, they seem the paradigm of what it sought to prohibit. Cf. Morris, 928 F.2d at 510 (access gained by guessing someone else’s password is not “authorization” under the Computer Fraud and Abuse Act).
The subpoena power is a substantial delegation of authority to private parties, and those who invoke it have a grave responsibility to ensure it is not abused. Informing the person served of his right to object is a good start, see Fed.R.Civ.P. 45(a)(1)(D), but it is no substitute for the exercise of independent judgment about the subpoena’s reasonableness. Fighting a subpoena in court is not cheap, and many *1075may be cowed into compliance with even overbroad subpoenas, especially if they are not represented by counsel or have no personal interest at stake. Because defendants procured consent by exploiting a mistake of which they had constructive knowledge, the district court erred by dismissing based on that consent.
Defendants ask us to affirm on the alternative ground that the messages they accessed were not in “electronic storage” and therefore fell outside the Stored Communications Act’s coverage. See 18 U.S.C. § 2701(a)(1). The Act defines “electronic storage” as “(A) any temporary, intermediate storage of a wire or electronic communication incidental to the electronic transmission thereof; and (B) any storage of such communication by an electronic communication service for purposes of backup protection of such communication.” Id. § 2510(17), incorporated by id. § 2711(1). Several courts have held that subsection (A) covers e-mail messages stored on an ISP’s server pending delivery to the recipient. See In re DoublecClick, Inc. Privacy Litig., 154 F.Supp.2d 497, 511-12 (S.D.N.Y.2001); Fraser v. Nationwide Mut. Ins. Co., 135 F.Supp.2d 623, 635-36 (E.D.Pa.2001);. cf. Steve Jackson Games, Inc. v. U.S. Secret Serv., 36 F.3d 457, 461-62 (5th Cir.1994) (messages stored on a BBS pending delivery). Because subsection (A) applies only to messages in “temporary, intermediate storage,” however, these courts have limited that subsection’s coverage to messages not yet delivered to their intended recipient. See DoubleClick, 154 F.Supp.2d at 512; Fraser, 135 F.Supp.2d at 636.
Defendants point to these cases and argue that messages remaining on an ISP’s server after delivery no longer fall within the Act’s coverage. But, even if such messages are not within the purview of subsection (A), they do fit comfortably within subsection (B). There is no dispute that messages remaining on NetGate’s server after delivery are stored “by an electronic communication service” within the meaning of 18 U.S.C. § 2510(17)(B). Cf. DoubleClick, 154 F.Supp.2d at 511 (holding that subsection (B) did not apply because the communications at issue were not being stored by an electronic communication service). The only issue, then, is whether the messages are stored “for purposes of backup protection.” 18 U.S.C. § 2510(17)(B). We think that, within the ordinary meaning of those terms, they are.
An obvious purpose for storing a message on an ISP’s server after delivery is to provide a second copy of the message in the event that the user needs to download it again' — if, for example, the message is accidentally erased from the user’s own computer. The ISP copy of the message functions as a “backup” for the user. Notably, nothing in the Act requires that the backup protection be for the benefit of the ISP rather than the user. Storage under these circumstances thus literally falls within the statutory definition.3
One district court reached a contrary conclusion, holding that “backup protection” includes only temporary backup storage pending delivery, and not any form of “post-transmission storage.” See Fraser, 135 F.Supp.2d at 633-34, 636. We reject this view as contrary to the plain language of the Act. In contrast to subsection (A), subsection (B) does not distinguish between intermediate and post-transmission storage. Indeed, Fraser’s interpretation renders subsection (B) essentially super*1076fluous, since temporary backup storage pending transmission would already seem to qualify as “temporary, intermediate storage” within the meaning of subsection (A). By its plain terms, subsection (B) applies to backup storage regardless of whether it is intermediate or post-transmission.
The United States, as amicus curiae, disputes our interpretation. It first argues that, because subsection (B) refers to “any storage of such communication,” it applies only to backup copies of messages that are themselves in temporary, intermediate storage under subsection (A). The text of the statute, however, does not support this reading. Subsection (A) identifies a type of communication (“a wire or electronic communication”) and a type of storage (“temporary, intermediate storage ... incidental to the electronic transmission thereof’). The phrase “such communication” in subsection (B) does not, as a matter of grammar, reference attributes of the type of storage defined in subsection (A). The government’s argument would be correct if subsection (B) referred to “a communication in such storage,” or if subsection (A) referred to a communication in temporary, intermediate storage rather than temporary, intermediate storage of a communication. However, as the statute is written, “such communication” is nothing more than shorthand for “a wire or electronic communication.”
The government’s contrary interpretation suffers from the same flaw as Fraser' s: It drains subsection (B) of independent content because virtually any backup of a subsection (A) message will itself qualify as a message in temporary, intermediate storage. The government counters that the statute requires only that the underlying message be temporary, not the backup. But the lifespan of a backup is necessarily tied to that of the underlying message. Where the underlying message has expired in the normal course, any copy is no longer performing any backup function. An ISP that kept permanent copies of temporary messages could not fairly be described as “backing up” those messages.
The United States also argues that we upset the structure of the Act by defining “electronic storage” so broadly as to be superfluous and by rendering irrelevant certain other provisions dealing with remote computing services. The first claim relies' on the argument that any copy of a message necessarily serves as a backup to the user, the service or both. But the mere fact that a copy could serve as a backup does not mean it is stored for that purpose. We see many instances where an ISP could hold messages not in electronic storage — -for example, e-mail sent to or from the ISP’s staff, or messages a user has flagged for deletion from the server. In both cases, the messages are not in temporary, intermediate storage, nor are they kept for any backup purpose.
Our interpretation also does not render irrelevant the more liberal access standards governing messages stored by remote computing services. See 18 U.S.C. §§ 2702(a)(2), 2703(b). The government’s premise is that a message stored by a remote computing service “solely for the purpose of providing storage or computer processing services to [the] subscriber,” id. §§ 2702(a)(2)(B), 2703(b)(2)(B), would also necessarily be stored for purposes of backup protection under section 2510(17)(B), and thus would be subject to the more stringent rules governing electronic storage. But not all remote computing services are also electronic communications services and, as to those that are not, section 2510(17)(B) is by its own terms inapplicable. The government notes that remote computing services and electronic communications services are “often the same entities,” but “often” is not good *1077enough to make the government’s point. Even as to remote computing services that are also electronic communications services, not all storage covered by sections 2702(a)(2)(B) and 2703(b)(2)(B) is also covered by section 2510(17)(B). A remote computing service might be the only place a user stores his messages; in that case, the messages are not stored for backup purposes.
Finally, the government invokes legislative history. It cites a passage from a 1986 report indicating that a committee intended that messages stored by a remote computing service would “continue to be covered by section 2702(a)(2)” if left on the server after user access. H.R.Rep. No. 647, 99th Cong., at 65 (1986). The cited discussion addresses provisions relating to remote computing services. We do not read it to address whether the electronic storage provisions also apply. See id. at 64-65. The committee’s statement that section 2702(a)(2) would “continue” to cover e-mail upon access supports our reading. If section 2702(a)(2) applies to e-mail even before access, the committee could not have been identifying an exclusive source of protection, since even the government concedes that unopened e-mail is protected by the electronic storage provisions.
The government also points to a subsequent, rejected amendment that would have made explicit the electronic storage definition’s coverage of opened e-mail. See H.R.Rep. No. 932, 106th Cong., at 7 (2000). This sort of legislative history has very little probative value; Congress might have rejected the amendment precisely because it thought the definition already applied.
Finally, the government points to a passing reference in another recent report. H.R.Rep. No. 236(1), 107th Cong., at 57 (2001). The discussion addresses an unrelated issue and merely assumes that the definition only applies to unopened e-mail. This assumption was natural given that Fraser had recently been decided.
We acknowledge that our interpretation of the Act differs from the government’s and do not lightly conclude that the government’s reading is erroneous. Nonetheless, for the reasons above, we think that prior access is irrelevant to whether the messages at issue were in electronic storage. Because plaintiffs email messages were in electronic storage regardless of whether they had been previously delivered, the district court’s decision cannot be affirmed on this alternative ground.4
2. Plaintiffs also claim a violation of the Wiretap Act, which authorizes suit against those who “intentionally intercepte ] ... any wire, oral, or electronic communication.” 18 U.S.C. §§ 2511(l)(a), 2520(a). We recently held in Konop v. Hawaiian Airlines, Inc., 302 F.3d 868 (9th Cir.2002), that the Act applies only to “acquisition contemporaneous with transmission.” Id. at 878. Specifically, “ ‘Congress did not intend for “intercept” to apply to “electronic communications” when those communications are in “electronic storage.” ’ ” Id. at 877 (quoting Steve Jackson Games, 36 F.3d at 462). Konop is disposi-*1078tive, and the district court correctly dismissed the claim.
3. Plaintiffs finally claim a violation of the Computer Fraud and Abuse Act, which provides a cause of action against one who, inter alia, “intentionally accesses a computer without authorization or exceeds authorized access, and thereby obtains ... information from any protected computer if the conduct involved an interstate or foreign communication.” 18 U.S.C. § 1030(a)(2)(C), (g). The conduct must involve one of five factors listed in 18 U.S.C. § 1030(a)(5)(B), which include a loss in excess of $5000. Id. § 1030(a)(5)(B)®, (g).5
The district court dismissed without leave to amend on the theory that the Act does not apply to unauthorized access of a third party’s computer. It also dismissed for failure to allege damages or loss, though it noted that this omission might be cured by amendment. Plaintiffs do not dispute the latter defect, but urge us to reverse as to the former ground so they can amend.
The district court erred by reading an ownership or control requirement into the Act. The' civil remedy extends to “[ajny person who suffers damage or loss by reason of a violation of this section.” 18 U.S.C. § 1030(g) (emphasis added). “ ‘[T]he word “any” has an expansive meaning, that is, “one or some indiscriminately of whatever kind.” ’ ” HUD v. Rucker, 535 U.S. 125, 131, 122 S.Ct. 1230, 152 L.Ed.2d 258 (2002) (quoting United States v. Gonzales, 520 U.S. 1, 5, 117 S.Ct. 1032, 137 L.Ed.2d 132 (1997)). Nothing in the provision’s language supports the district court’s restriction. Individuals other than the computer’s owner may be proximately harmed by unauthorized access, particularly if they have rights to data stored on it.
Defendants argue in the alternative that NetGate authorized their access. Our earlier discussion disposes of this defense. They further contend that any damages or loss plaintiffs suffered do not fall within the Act’s ambit. Because plaintiffs have not yet alleged the damages or loss they suffered, it would be premature to consider the argument.
4. Defendants contend they are immune from liability under the Noerr-Pennington doctrine, which exempts petitioning of public authorities from civil liability on First Amendment grounds. See Manistee Town Ctr. v. City of Glendale, 227 F.3d 1090, 1092 (9th Cir.2000); Kottle v. Northwest Kidney Ctrs., 146 F.3d 1056, 1059 (9th Cir.1998). Lawsuits are protected by the doctrine because they are essentially petitions to the courts for redress of grievances. See Cal. Motor Transp. Co. v. Trucking Unlimited, 404 U.S. 508, 510, 92 S.Ct. 609, 30 L.Ed.2d 642 (1972). The doctrine is typically invoked to immunize the act of petitioning itself— i.e., the filing of the lawsuit. But it has been extended to certain conduct “incidental to the prosecution of the suit,” for example, deciding whether to settle a claim. See Columbia Pictures Indus., Inc. v. Prof'l Real Estate Investors, Inc., 944 F.2d 1525, 1528-29 (9th Cir.1991), aff'd, 508 U.S. 49, 113 S.Ct. 1920, 123 L.Ed.2d 611 (1993). Defendants seize on this language from Columbia Pictures and argue that, because the subpoena was incidental to their litigation, they are entitled to immunity.
We are skeptical that Noerr-Pennington applies at all to the type of con*1079duct at issue. Subpoenaing private parties in connection with private commercial litigation bears little resemblance to the sort of governmental petitioning the doctrine is designed to protect. Nevertheless, assuming arguendo the defense is available, it fails. Noerr-Pennington does not protect “objectively baseless” sham litigation. See Prof'l Real Estate Investors, Inc. v. Columbia Pictures Indus., Inc., 508 U.S. 49, 60, 113 S.Ct. 1920, 123 L.Ed.2d 611 (1993). The magistrate judge found that the subpoena was “transparently and egregiously” overbroad and that defendants acted with gross negligence and in bad faith. This is tantamount to a finding that the subpoena was objectively baseless.
Defendants urge us to look only at the merits of the underlying litigation, not at the subpoena. They apparently think a litigant should have immunity for any and all discovery abuses so long as his lawsuit has some merit. Not surprisingly, they offer no authority for that implausible proposition. Assuming Noerr-Pennington applies at all, we hold that it is no bar where the challenged discovery conduct itself is objectively baseless.6
5. Having dismissed all the federal claims, the district court declined jurisdiction over the pendent state law claims under 28 U.S.C. § 1367(c)(3). Because we reverse dismissal of some of the federal claims, we also reinstate the state claims.
We REVERSE dismissal of the Stored Communications Act claim, AFFIRM dismissal of the Wiretap Act claim, and REVERSE dismissal with prejudice of the Computer Fraud and Abuse Act claim with instructions to dismiss with leave to amend. We also REVERSE dismissal of the state claims.
AFFIRMED in part, REVERSED in part and remanded. Costs to appellants.
Defendants urge us to affirm on the additional ground that the complaint at one point alleges they "access[ed] ... Plaintiffs' computer systems through which electronic communications systems are provided.” Resp. Br. of Def./Appellee at 16 (emphasis added). Plaintiffs' computers, they note, are not "facilities” covered by 18 U.S.C. § 2701(a)(1). Id. at 15. We construe complaints liberally, however, see Lynn v. Sheet Metal Workers’ Int'l Ass’n, 804 F.2d 1472, 1482 (9th Cir.1986), and the substance of plaintiffs’ claims is that defendants improperly accessed NetGate’s servers.
. Defendants had a full and fair opportunity to litigate the validity and good faith of the subpoena in the sanctions proceedings. Those proceedings were filed as a miscellaneous action on a separate docket from both the underlying New York litigation and the instant suit; defendants did not appeal. The magistrate's findings are therefore preclusive. Cf. 18 James Wm. Moore et al., Moore's Federal Practice § 132.03[4][k][vii], at 132-126 (3d ed.1997) (judgment of contempt is issue pre-clusive); 3 Ronald E. Mallen & Jeffrey M. Smith, Legal Malpractice § 21.14, at 286 (5th ed.2000) (attorney disciplinary proceedings may be issue preclusive).
. Prosser and Keeton say that a plaintiff's consent is probably invalid if the defendant "ought to have known in the exercise of reasonable care” about the mistake. Prosser & Keeton § 18, at 119. Because the Stored Communications Act defines a criminal offense and includes an explicit mens rea requirement, see 18 U.S.C. § 2701(a)(1), we do not think a defendant can be charged with constructive knowledge on a showing of mere negligence. Rather, the defendant must have consciously procured consent through improper means. In this case, the magistrate found that defendants had acted in bad faith. That is enough to charge them with knowledge of NetGate's mistake. See Black's Law Dictionary 139 (6th ed.1990) (defining "bad faith” as "not simply bad judgment or negligence, but ... conscious doing of a wrong because of dishonest purpose or moral obliquity”).
. That defendants did not read the messages until NetGate posted them to a website is immaterial. Defendants' unlawful subpoena caused NetGate to retrieve the messages from electronic storage and make them available. That constitutes "access” within the meaning of the Act.
. Defendants urge us to affirm on the additional ground that the complaint at one point alleges they “access[ed] ... Plaintiffs’ computer systems through which electronic communications systems are provided.” Resp. Br. of Def./Appellee at 16 (emphasis added). Plaintiffs' computers, they note, are not "facilities” covered by 18 U.S.C. § 2701(a)(1). Id. at 15. We construe complaints liberally, however, see Lynn v. Sheet Metal Workers' Int’l Ass’n, 804 F.2d 1472, 1482 (9th Cir. 1986), and the substance of plaintiffs' claims is that defendants improperly accessed Net-Gate’s servers.
. Defendants argue that subsection (a)(5)(A) prescribes the Act’s only civil offenses. But subsection (g) applies to any violation of "this section” and, while the offense must involve one of the five factors in (a)(5)(B), it need not be one of the three offenses in (a)(5)(A).
. Noerr-Pennington s sham litigation exception also requires proof of subjective intent to use legal process to achieve the evil prohibited by the statute from which exemption is claimed. See Prof'l Real Estate Investors, 508 U.S. at 60-61, 113 S.Ct. 1920. That prong of the test is also satisfied here; presumably, the purpose of any objectively baseless subpoena is to uncover private information.