From advanced encryption to large-scale testing, a Swiss prototype reopens the debate on e-voting as a trusted democratic infrastructure.

In recent years the debate onintegrity of electoral processes has intensified significantly. In many areas of the world, the trust in voting mechanisms is put under pressure by phenomena of misinformation, external interference and practices of voter intimidation. In this context, the promise of the remote electronic voting continues to appear seductive to governments and institutions: greater accessibility, reduced logistical costs, potential increase in attendance. However, a structural obstacle has so far limited its large-scale adoption: resistance to coercion, that is, the ability to ensure that no voter can be forced or incentivized to vote in a certain way.

A recent prototype developed at the Federal Polytechnic of Lausanne It now seems to indicate that this limit is no longer insurmountable. The Swiss system, called Votegral, represents one of the most advanced and concrete attempts to transform electronic voting from a theoretical exercise to a workable democratic infrastructure, capable of competing with traditional models based on paper ballots and voting booths.

Why coercion remains the critical issue in digital voting

Voting in person, with identification of the voter, closed booth and physical deposit of the ballot, is still considered the more solid standard in terms of protection against external pressure. It is not a perfect system, but it makes it extremely difficult to prove to third parties how one voted, thus reducing the risk of vote buying or systematic intimidation.

In remote electronic voting, this guarantee disappears. A voter may find himself voting under the gaze of a family member, an employer or, in more extreme scenarios, of organized actors who offer financial compensation in exchange for proof of the vote cast. This is what makes the coercion resistance one of the most complex requirements to satisfy in a digital environment. Many e-voting systems tested over the last twenty years have privileged computer security and anonymity, without completely eliminating the possibility of demonstrating one's vote to a coercer.

Votegral is the idea of ​​a robust end-to-end system

The project Votegral is born within the Decentralized and Distributed Systems Laboratory (DEDIS) of the EPFL and is proposed as complete e-voting pipeline, from voter registration to the final count. The distinctive element is the systematic use of fake credentials Voluntarily generated by voters under duress. In practice, a citizen can produce a seemingly valid credential and use it to submit a "fake" vote, which will be automatically discarded during the final count.

The system is based on zero-knowledge cryptographic proofs interactive, which allow the correctness of the operations to be verified without revealing the identity of the voter or the content of the vote. The combination of these tools allows for break the link between voter and ballot more radically than previous solutions.

“Votegral guarantees that no voter can prove how they voted, even if they wanted to,”

has explained Bryan Ford, professor and head of the DEDIS laboratory at EPFL.

“This renders both vote-buying and forms of domestic or organized coercion ineffective.”

From theoretical limits to computational feasibility

One of the reasons why similar approaches never made it out of the lab was the computational costIn previous systems, the time required to count large volumes of ballots was simply incompatible with any real election. In some cases, counting a million votes would have taken centuries.

the prototype Votegral has demonstrated that this bottleneck can be overcome. Thanks to a series of optimizations in the generation of cryptographic proofs and in the validation phase filtering fake credentials, the system managed to complete the counting of one million votes in approximately 14 hoursA result that places Votegral within the time limits set by the electoral legislation of many countries.

Secondo Louis-Henri Merino, PhD student at DEDIS and first author of the study presented at Symposium on Operating Systems Principles 2025,

“The performance achieved is competitive with the e-voting systems already in use in Switzerland and demonstrates that resistance to coercion is no longer a theoretical luxury, but a concrete possibility.”

Competitiveness and integration in sistemi esistenti

Another relevant aspect of the project is its modular architecture. Votegral has been designed to be compatible with electronic voting infrastructure already operational, such as those used by Swiss Post in some cantonal contexts. In perspective, the system could be integrated as additional level of security, without requiring a complete rewrite of existing platforms.

This design choice opens up to international collaboration scenarios and reduces barriers to adoption. It's not about replacing current models overnight, but about strengthen them with a mechanism that addresses one of the most critical points of remote voting.

The political factor and the challenge of the quantum future

If from a technical point of view the problem of coercion appears to be manageable today, questions of political and institutional nature. Many governments have so far avoided electronic voting not only for technological reasons, but also because of the difficulty of publicly verify the code and to explain complex cryptographic systems to citizens.

Added to this is an emerging concern: the impact of the quantum calculation on current cryptography. The possibility that future quantum computers could decrypt data currently considered secure raises questions about retroactive protection of the secrecy of the voteVotegral researchers argue that conceptual solutions already exist for a post-quantum version of the system, although this step will require further years of development.

A paradigm shift in the e-voting debate

The value of Votegral goes beyond the single prototype. The project contributes to shifting the debate from the idea that the resistance to coercion is incompatible with digital voting, in the knowledge that it is a solvable technical problem, under certain conditions. As he observed Bryan Ford,

“Research has shown that the problem can be solved end-to-end; what is missing now is the political will to transform this solution into public infrastructure.”

In an era where the trust in democratic processes is increasingly fragile, initiatives such as Votegral indicate that thetechnological innovation It can still play a structural role, not as a shortcut, but as a rigorous tool at the service of institutions. The challenge is no longer to demonstrate that electronic voting can work, but to decide whether and how to responsibly integrate it into existing systems. future democratic systems.

Here are three insights that might interest you:

Liberland, the Blockchain State Where Freedom is Programmable
Eligo Next is the new Italian electronic voting platform
How to safeguard innovation from the high cost of money?