Ensuring the integrity of paper ballots is a pressing concern for election officials and advocates. Traditional safeguards like chain-of-custody protocols and serial numbers offer some protection, but emerging research from Princeton University is focused on "paper fingerprinting", using the random, unique patterns in paper fibers or embedded UV flakes to authenticate ballots.
Limitations of High-Tech Watermarks
While high-tech watermarks, including ultraviolet (UV) markings and holograms, appear impressive, they are not as foolproof as they seem. The blog points out that custom stamps, UV inks, and commercial printing with UV phosphorescent ink are widely available and affordable.
This accessibility means determined attackers could replicate these security features, especially when special paper is distributed to thousands of jurisdictions, reducing its uniqueness and effectiveness.
The Power of Unique Paper Patterns
More robust alternatives exist, such as leveraging the unique, random fiber patterns found in ordinary paper. Research demonstrates that these patterns are distinct and can be captured with high-resolution scanners, effectively making each ballot its own "fingerprint."
This concept extends to paper embedded with random UV flakes, which can create natural signatures that are extremely difficult to duplicate at scale.
Integrating Fingerprints in Ballot Tracking
Many U.S. elections already use optical-scan machines that generate a cast vote record (CVR) for each ballot, sometimes linking it to a serial number.
The proposed innovation is for these machines to also scan and record each ballot's paper fingerprint, storing this data in the CVR.
This would provide each ballot with a virtually unforgeable identifier, significantly enhancing audit reliability and reducing the risk of fraudulent ballots entering the system.
Unlike visible serial numbers, paper fingerprints are much harder to forge. They could also strengthen chain-of-custody tracking, ensuring ballots match their CVR records with high confidence. However, it's crucial that fingerprints are scanned at the point of voting to preserve the secret ballot and prevent linking votes to individual voters.
Challenges and Questions
- Replication Risk: Can the random patterns in UV-flake paper truly not be replicated at scale? This claim needs scientific validation, not just marketing assurances.
- Voter Privacy: Pre-scanning and tracking ballot fingerprints before voting could undermine the secret ballot, risking coercion or vote-buying.
- Technology Reliability: The system's security depends on computers accurately generating fingerprint numbers from physical patterns. If compromised, these systems could undermine the benefits, especially since fingerprints are not human-readable and require trust in the technology.
- Implementation Complexity: Transitioning to paper fingerprinting would demand new protocols, reliable hardware and software, and comprehensive security analysis. Pilot projects are essential before considering nationwide adoption.
Comparing to Current Approaches
While fingerprinting could offer enhanced security, especially against advanced tampering, its added value may not justify the complexity of overhauling current election infrastructure.
Limited pilot programs could help determine if the benefits of paper fingerprinting are significant enough to warrant broader use.
Paper fingerprinting is a promising yet unproven tool for election integrity. While it could strengthen ballot tracking and auditing, serious challenges, ranging from replication risks to privacy and technological reliability, must be addressed.
For now, cautious experimentation and careful evaluation are recommended over immediate, widespread adoption.
Could Paper Fingerprinting Transform Ballot Security?