The integrity of electoral systems in the United States raises ongoing questions regarding cybersecurity, including the security of voting machines, counting mechanisms, and the risks of digital influence on voters. Despite the existence of multiple preventive measures, experts believe that the technical and operational aspects of elections face complex challenges that necessitate a comprehensive reconsideration of the standards of trust placed in the current infrastructure.
Technical Challenges in Voting and Counting Machines
Many electoral districts rely on ballot-marking devices to print a ballot that includes the voter's choice and a human-readable element, alongside a QR code or barcode that is scanned to count the votes. Although these devices receive certification from the Election Assistance Commission (EAC) and are subject to the air-gapped principle, experts point to potential vulnerabilities in the maintenance and update cycle.
Shay Perdan, co-founder and CEO of Sequent, explains that the certification process for voting machines can be time-consuming, which may leave software without regular patch updates for many years compared to practices followed in the commercial technology sector. Furthermore, field penetration tests in election security labs, such as the Voting Village at the DEF CON conference, have indicated that devices available to researchers can be compromised when direct physical access is provided.
Risks of Influence Campaigns and Artificial Intelligence
Beyond hardware infrastructure, cyber disinformation campaigns and the manipulation of public awareness pose a major challenge. Adam Marre, a former cybersecurity official at the FBI, explains that geopolitical actors typically aim to undermine trust in democratic institutions and influence political narratives rather than attempting to directly alter votes within the systems.
With the development of artificial intelligence tools, attacks have become more scalable and personalized. Early signs of this have emerged through deepfake technologies, such as automated calls that mimicked the voices of prominent political figures with the goal of discouraging voters from participating. Experts warn that exploiting artificial intelligence to promote targeted rumors against specific candidates could sway results in highly polarized communities.
What is Changing Practically and What are the Proposed Solutions?
There is no physical evidence to date confirming that the results of any US election have been actually manipulated, thanks to oversight measures and multiple manual audits of paper ballots. However, specialists suggest technical steps to enhance reliability without the need for a radical change to the electoral process:
- Open-source software: Allowing academics and cybersecurity experts to review the source code of voting machines using analysis tools, penetration tests, and artificial intelligence to detect issues early.
- Homomorphic encryption for results: Using advanced encryption techniques such as homomorphic encryption to create independent digital voting records that provide mathematical proof and independent verifiability of the count's accuracy alongside the traditional paper ballot.
- Protecting the surrounding operational layer: Securing voter registration databases, county websites, and the email accounts of campaign workers, as they are the first line of defense against external hacking attempts.
Ultimately, experts emphasize that achieving absolute security in the current digital and geopolitical environment is extremely difficult, making reliance on transparency, continuous auditing, and the protection of societal trust essential pillars alongside technology itself.