Oral arguments for the next round in United States v. Nosal have been set for October 20, 2015 at the Ninth Circuit in San Francisco. So we figured it may be a good time to review both sides’ arguments related to the Computer Fraud and Abuse Act. After doing so, it seems to us that one topic not given any consideration in the briefs, but that may play a role during oral argument is the phenomenon known as phishing schemes, and how such schemes might be compared and contrasted with the scheme alleged in this case.
Background
As covered previously by Trade Secrets Watch, this closely-watched case involves the prosecution of David Nosal, a former executive with the recruitment and talent management firm, Korn Ferry. Nosal was originally charged with eight counts of violating the CFAA in 2008 for conspiring with then-current and former Korn Ferry employees to obtain confidential data from the firm’s “Searcher” database, which he used to start a competing venture.
In April 2012, an en banc panel of the Ninth Circuit (Nosal I) ruled that the government could not bring five of the eight CFAA counts against Nosal because he could not have “exceed[ed] authorized access” under the CFAA when his co-conspirators used their own valid log-in credentials to access and download confidential prospect lists from Korn Ferry’s database while still working at the firm.
The decision left unresolved three remaining CFAA counts against Nosal. These counts related to incidents in 2005 when two of Nosal’s alleged co-conspirators, who by then had left Korn Ferry, accessed the database using the login credentials of Nosal’s former secretary, who remained at the firm and voluntarily gave them her credentials. Unlike the “exceeding authorized access” counts dismissed in Nosal I, these counts charged that Nosal was “without authorization” – the second possible avenue for charging unauthorized access under the CFAA.
In April 2013, a jury convicted Nosal of the CFAA counts, as well as two counts of violating the Economic Espionage Act for theft of trade secrets. The district court stayed his 366-day sentence pending the current appeal.
Nosal’s CFAA Arguments
Nosal’s primary CFAA argument is that the en banc opinion in Nosal I held that the CFAA criminalizes hacking, which the panel defined as “the circumvention of technological access barriers—not the misappropriation of trade secrets.” Relying heavily on the reasoning from the en banc opinion, Nosal’s opening brief argues that password sharing is not hacking and should not be a federal crime. In particular, Nosal points to a passage in Nosal I where author Judge Alex Kozinksi reasons that “Facebook makes it a violation of the terms of service to let anyone log into your account. . . . but few imagine they might be marched off to federal prison for doing so.” Nosal’s brief chastises the district court for disregarding this part of the en banc opinion.
The brief then attempts to draw a line between a hacker who steals or guesses a victim’s password and uses it to gain access – something that Nosal says would be “without authorization” – and an authorized password holder who voluntarily shares his or her password with someone who is not authorized – which would not be “without authorization.”
This is a fine line indeed, particularly in a world where one common tactic of hackers is “phishing” – an attempt to obtain sensitive information such as usernames or passwords by tricking victims into divulging it. Significantly, in a phishing scheme, if the victim is tricked into giving the perpetrator his password, the transfer is nonetheless made voluntarily.
It would not be correct to characterize Nosal and his co-conspirators’ acquisition of his former secretary’s password as phishing. The secretary was apparently cooperating with Nosal and provided her password to his co-conspirators. However, one similarity between the situation in Nosal and a phishing scenario is that they both involve an insider voluntarily providing sensitive information to an outsider, allowing the outsider to gain access to a computer system. The main difference, of course, is the motivation of the insider: a typical phishing victim is innocent, whereas Nosal’s former secretary was essentially a co-conspirator. But from the perspective of a company – and potentially a court looking to determine the culpability of the outsider – the two circumstances look quite similar: an outsider gains access to sensitive data by convincing an insider to voluntarily provide access.
Nosal, therefore, appears to be vulnerable to an argument that, if the law in the Ninth Circuit really requires that in order to show “without authorization,” a defendant must circumvent a “technological access barrier,” that rule must implicate outsiders whenever an insider voluntarily assists with the circumvention, whether that insider is an innocent victim of trickery or a co-conspirator. To hinge an outsider’s CFAA liability on the mindset of the insider who provides access would not only create a more complicated rule, but potentially allow one set of clearly-culpable outsiders to escape liability while another set incurs liability, when both are engaging in similarly illicit conduct – using an insider’s credentials to gain unauthorized access to a computer system.
While the government’s brief does not directly employ the phishing argument, it comes close.
The Government’s CFAA Arguments
The government argues simply that by password-protecting its computers, Korn Ferry did create a “technological access barrier” intended to prevent access by non-employees. Accordingly, circumvention of that barrier, even via the voluntary transfer of a password from an authorized employee to an outsider, should be deemed to violate the CFAA. The government also points out that Section (a)(6) of the CFAA, which prohibits fraudulent trafficking in passwords, expressly contemplates that use of another person’s password to access a computer may be considered “without authorization.”
However, the government’s primary argument is that Nosal I actually did not hold that the “the circumvention of a technological access barrier” is a required component for all CFAA cases, and to the extent that it is a requirement at all, it is only applicable to “exceeding authorized access” – the element at issue in the first case, not this one.
The rule regarding “without authorization,” according to the government, is governed by the Ninth Circuit’s pre-Nosal holding in LVRC Holdings LLC v. Brekka. In Brekka, a plaintiff company sued a former employee for emailing proprietary company documents to himself and his wife while still employed by the company. It claimed the defendant emailed the documents “to further his own personal interests, rather than the interests of [the company],” and was thus “without authorization.” The panel rejected the so-called agency approach adopted by the Seventh Circuit in International Airport Centers v. Citrin and found the defendant was not liable to his employer under the CFAA, because the employer had not yet revoked the defendant’s access to their computers.
Despite Brekka’s pro-defendant outcome, the government’s brief capitalizes on the opinion’s explanation of the circumstances where a person can be “without authorization.” One such circumstance is “when the employer has rescinded permission to access the computer and the defendant uses the computer anyway.” This, according to the government, precisely fits the actions of Nosal’s co-conspirators.
Phishing Could Be a Factor
While the government correctly points out that Nosal I expressly did not overrule Brekka, the government’s primary obstacle is that the reasoning in Nosal I would appear to apply to access “without authorization” just as much as it does to “exceeding authorized access.” In other words, if Nosal I did not overrule the holding in Brekka, it appears to have at least refined it. While the actual holding of Nosal I is in dispute, the result unquestionably turned in large part on the panel’s concerns that the CFAA could be interpreted to criminalize otherwise innocuous behavior that merely violates employer use restrictions or a website’s terms of service.
The government’s brief attempts to address these concerns, in part, by calling hypotheticals, such as the prosecution of cases involving password sharing between loved ones, “unlikely.” However, the majority opinion in Nosal I made it clear that it would not take the government’s word for it. To make that point, Judge Kozinski cited United States v. Drew, a case where the government charged a woman with violating the CFAA for setting up a fictitious MySpace profile in violation of the website’s terms of service.
A more fruitful path for the government may be to draw on the similarities between Nosal’s scheme and phishing schemes. Nosal’s counsel will be able to distinguish the two. However, if the requirement for “circumvention of a technological access barrier” was indeed part of the holding in Nosal I, it may be awkward to have a rule that implicates the CFAA when circumvention stems from phishing but not when it stems from a willing insider. Presented with the choice between creating a rule that might be interpreted to allow phishers to pass through the CFAA’s net on the one hand and affirming Nosal’s conviction on the other, the Ninth Circuit might settle on the latter. The Panel could, for example, simply hold that the requirement for “circumvention of a technological access barrier” was the holding in Nosal I, and that by obtaining login credentials from Nosal’s former secretary, Nosal and his co-conspirators achieved the necessary circumvention.
We’ll be watching.