Is Big Tech a big election security risk?
- By Abel Morales
- Oct 29, 2020
With the U.S. general election swiftly approaching, advocacy groups are directing their attention to a very specific vector for potential voter data compromise: Big Tech. According to CBS News, over the course of the Trump presidency, a growing number of state and local governments have turned to Amazon Web Services to store and host voter information, a number that’s likely only risen since the pandemic created an increased reliance on the cloud. These groups, which include Colors of Change, Demands Progress and RootsAction, have every right to be concerned.
Cloud-based voter data compromises
In 2017, security researchers discovered that detailed information on 198 million voters in the U.S. were left exposed online. The files belonged to a political consultancy working for the Republican National Committee and contained names, addresses, birth dates, phone numbers and detailed personal voting information pulled from previous voting records and online activity. A year earlier, a similar event – exposure of an unsecured database – potentially compromised the identity of 87 million Mexican voters. Amazon has cited customer error as the reason for these data exposures.
These AWS exposures along with other instances of voter data harvesting and misuse of voter information like the Facebook-Cambridge Analytica data scandal have raised concerns about Big Tech’s data protection responsibilities. Advocacy groups are questioning whether companies that provide cloud services are doing enough to safeguard voter data and prevent interference in the 2020 election.
An increasing number of campaigns and consultant firms are using cloud services to build more detailed profiles of voters. The hosting or web services offered by cloud providers also store election results, support campaign websites and voter registration databases. With these cloud providers responsible for storing so much sensitive information, coupled with an increase in credential theft as a result of our current remote work environment, both Big Tech and the government must address the major risks associated with compromised credentials to preserve election security.
Why credential threats should be top priorities
This year’s Verizon Data Breach Investigations Report (DBIR) revealed more than 80% of breaches tied to hacking involve the use of lost or stolen credentials or those obtained by brute force. Compromised credentials allow attackers to laterally move through the network -- they escalate their privileges to obtain access and exfiltrate critical data to sell on the dark web or be used later for more sophisticated attacks. Credential-based attacks on Google, Yahoo and Microsoft in 2019 allowed attackers to access user information, further highlighting the need for Big Tech to prevent attacks on their cloud databases storing critical voter information.
Despite rising concerns, reliance on the cloud will continue to grow as a result of the pandemic and a general positive trajectory towards cloud usage. Organizations using cloud providers for election data must take active measures to protect their sensitive information. Simple and effective measures involve investing in education around security hygiene and best practices for password security; implementing password managers; constantly monitoring the user credentials of employees, volunteers and third parties for weak or stolen passwords; centralizing cloud logs with on-premise logs for a true, single-pane-of-glass visibility; utilizing biometrics where possible; and enforcing the use of multi-factor authentication everywhere.
Additionally, government agencies can further secure the data in their cloud by implementing cloud-based SIEM and behavioral analytics. User and entity behavior analytics uses machine learning to determine normal and abnormal behavior by users, machines and other entities on the network. This helps to spot potential attacks early on and identify the risky activity, even if it has never been seen before, and spans multiple organizational systems and data sources. This means government organizations can reduce the resources spent mitigating and responding to cyberattacks and protect their clouds with analytics capable of analyzing access and authentication data, establishing user context and reporting on suspicious activity.
Big Tech and government agencies must team up
While the onus to protect the data falls on the customer, the amount of sensitive information residing in the cloud means Big Tech is responsible for disclosing data breaches. Government agencies can use that knowledge in conjunction with cloud-based security solutions to protect voter data and their own reputations.
As we look to the future of election security, Big Tech and government must work together to safeguard our elections and credentials while enabling organizations to safely collect and store necessary data to inform elections-based decisions. The combined efforts of Big Tech and government will increase the nation’s ability to protect elections infrastructure and citizen data during this election cycle and beyond.
Abel Morales is a security engineer at Exabeam.