Organizations at risk of a data breach (that’s every organization, by the way) can learn something from Uber’s data privacy missteps. Squire Patton Boggs attorneys Colin Jennings, Ericka Johnson and Dylan Yépez offer key takeaways from the company’s high-profile data breaches.
On August 19, 2020, the former Chief Security Officer (CSO) for Uber Technologies Inc. (Uber) was charged with obstruction of justice and misprision of felony for allegedly trying to conceal from federal investigators a cyberattack that occurred in 2016, exposing the data of 57 million riders and drivers. Although an extreme case, it is a good reminder for companies and executives to take data breach disclosure obligations seriously.
The criminal complaint, filed in the U.S. District Court for the Northern District of California (“the Complaint”), appears to claim that Uber, through its former CSO, Joseph Sullivan, should have reported the 2016 data breach to federal investigators. But a business’s duty to disclose a data breach is not always clear, and there are often a myriad of laws, regulatory practices and consumer expectations when navigating a breach. Using Uber’s 2016 breach as a case study, company executives must be aware of and recognize the business and personal consequences associated with breach response, and specifically with intentionally concealing a breach.
The Obligation to Report a Data Breach is Often Not Straightforward
Across the world, countries have widely varying laws related to the protection of personal information and even greater variance on the requirements to disclose a breach of such information. Even within the United States, the definitions of “personal information” and “data breach” differ greatly from state to state, with no two state laws being identical, so businesses, particularly those operating on a national or global scale, must conduct multijurisdictional analyses to determine whether an obligation to disclose a given breach exists and, if so, the scope of the obligation. Often there are inconsistent laws and obligations, and regulatory and consumer expectations can vary greatly based on the nature, scope and context of the breach.
Many laws require disclosure of a data breach only if there is a “reasonable risk of harm” to the individual(s) whose personal information was unlawfully accessed and/or exfiltrated. This requires businesses to determine whether, based on the totality of circumstances, it is reasonably likely that a breach of personal information will harm affected individuals. On the other hand, some laws do not require any risk of harm. Further, given that the forensic review of a data breach evolves over time, it is not uncommon for the initial findings to change dramatically over the course of a breach response. What often appears to be a limited attack can become a wholesale loss of sensitive consumer or business data – and oftentimes both simultaneously.
The legal analysis is then complex, fact-specific and ever changing. Perhaps, for example, only a portion of the sensitive data was exposed (e.g., only the last four digits of a social security number or only an individual’s last name). Maybe, due to insufficient logs, forensic investigators cannot rule out the possibility that an unauthorized third party accessed the sensitive data or moved laterally into human resources data or databases containing consumer financial information. Or perhaps evidence suggests that the cybercriminals appear to be staging sensitive data for exfiltration, but have destroyed any evidence that data was actually taken. These are but a few examples of factors that can make the obligation to report far from straightforward.
As Uber’s 2016 breach response indicates, the difficulty of ascertaining a business’s breach notification obligations is not a defense to those company executives who intentionally conceal a breach. As discussed below, company executives who ultimately have to decide whether to disclose a breach should take notice of the potential consequences of making the wrong decision.
A Case Study in Intentionally Failing to Report a Breach
The Complaint alleges that, in response to Uber’s 2016 breach, former CSO Joseph Sullivan “engaged in a scheme to withhold and conceal from the [Federal Trade Commission] both the hack itself and the fact that that data breach had resulted in the hackers obtaining millions of records associated with Uber’s users and drivers.”
At the time of the breach, Sullivan was helping oversee Uber’s response to a Federal Trade Commission (FTC) investigation into Uber’s data security practices, which had been triggered, in part, by another Uber data breach that occurred in or around 2014. Sullivan was “intimately familiar with the nature and scope of the FTC’s investigation.”
About 10 days after providing sworn testimony to the FTC, however, Sullivan received an email from “johndoughs@protonmail.com,” claiming to have found a “major vulnerability in uber [sic],” and threatening that the hacker “was able to dump uber [sic] database and many other things.” Within days, Sullivan’s security team realized that an unauthorized person or persons had accessed Uber’s data and obtained, among other things, a copy of a database containing approximately 600,000 driver’s license numbers for Uber drivers.
Based on available information, this massive data breach likely triggered Uber’s duty to notify under numerous jurisdictions’ data breach laws. By contrast, the 2016 breach appeared significantly more expansive than the 2014 breach, in which a cybercriminal accessed over 100,000 individuals’ personal information on a cloud-based data warehouse.
Based on the Complaint, Sullivan allegedly took affirmative measures to conceal the data breach and the resulting exposure of data. Among other things, he allegedly:
- negotiated with the cybercriminals to pay $100,000 in exchange for the hackers to sign a nondisclosure agreement (NDA), “falsely represent[ing] that the hackers had not obtained or stored any data during their intrusion,” even though “[b]oth the hackers and Sullivan knew at the time that this representation in the NDA was false;”
- “instructed his team to keep knowledge of the 2016 breach tightly controlled;”
- “never informed the FTC of the 2016 data breach, even though he was aware that the FTC’s investigation focused on data security, data breaches and protection of [Personally Identifiable Information];” and
- “removed certain details … that would have illustrated the true scope of the [2016] breach” from a prepared summary for the new Uber CEO – changes which “resulted in both affirmative misrepresentations and misleading omissions of fact.”
Sullivan’s alleged motives to cover up the 2016 hack and data breach are the concerns that all companies must assess in connection with their breach notification responsibilities.
First, the Complaint appears to allege that one motive to conceal the breach was to prevent further reputational harm to the company. Like Uber’s customers, individuals entrust their data to companies on a daily basis, from making purchases to requesting services. Companies know, therefore, that they risk losing revenue if their customers lose confidence in the protection of their data.
Understanding this dynamic, he “became aware the attackers had accessed [the cloud] in almost the identical manner the 2014 attacker had used,” according to the Complaint. “That is, the attackers were able to access Uber’s source code on GitHub (this time by using stolen credentials), locate [a cloud] credential and use that credential to download Uber’s data.” As such, the Complaint appears to allege that both the embarrassment of falling victim to the same attack vector and the associated reputational consequences may have motivated Sullivan to conceal the breach.
Second, the Complaint appears to allege that another motive for concealing the breach was to prevent additional regulatory scrutiny. In the United States, companies like Uber are subject to many state- and industry-specific regulators (e.g., state Attorneys General, the Securities and Exchange Commission, FTC) — often simultaneously. Additionally, outside of the United States there are numerous laws and data protection or other authorities that govern data breaches.
At the time of the breach, Sullivan was actively responding to the FTC’s inquiries to assist in reaching a settlement related to the 2014 breach. For example, he approved language to the FTC representing that “‘all new database backup files’ had been encrypted since August 2014,” when in fact, they had not. Sullivan’s fears may not have been misplaced. In light of the new information regarding the 2016 breach, the FTC effectively withdrew its previous settlement terms and added requirements to the resolution with Uber.
Ultimately, it appears that such attempts to rationalize and avoid Uber’s breach notification responsibilities may have led Sullivan to engage in the actions he did.
Lesson Learned
In a public statement, the FBI advised that, “[w]hile this case is an extreme example of a prolonged attempt to subvert law enforcement, we hope companies stand up and take notice.” In effect, the consequences of failing to disclose a data breach are the most extreme in cases where a notification obligation clearly exists and the company and its officers consciously decide to circumvent that obligation during the course of an ongoing investigation. While companies have incentives to rationalize and avoid their disclosure obligations (e.g., reputational harm, regulatory oversight, expense), this incident highlights the potential consequences executives should be aware of when weighing the business decision to disclose a breach. Disclosure and direct individual notification of a data breach is now the expectation, and the decision to not disclose must be very carefully weighed – taking into account law, regulatory practice and consumer/customer expectations. One size does not fit all, and the nature, scope and circumstance of the specific breach must be carefully assessed in real time.
Ultimately, the legal analysis to determine whether an obligation exists and the business decision to disclose the same are nuanced and complex. If you experience a data breach, it is best to retain counsel who is highly experienced in the nuances of data breaches and the complexities of data breach notification laws for help determining whether and how to disclose a given breach.