[New Wave] Is Voting Using Blockchain Really Safe?

Blockchain has the characteristic of 'decentralization,' where data is transparently disclosed to all participants, and when processing this data, consensus is reached by asking all participants for their opinions. However, blockchain was not the first to propose this consensus-based data processing method; many cryptographers had already published various consensus algorithms long before.

The greatest academic achievement of blockchain is that, unlike existing consensus algorithms which require participants to be predetermined, the blockchain proposed by Satoshi Nakamoto has no limit on the number of participants, allowing users to freely join or leave the consensus at any time. This is called 'permissionless consensus' or 'Nakamoto consensus,' and the foundation that enables stable consensus despite such free participation or withdrawal is the incentive provided through Bitcoin. It is worth noting that articles often mistakenly introduce 'Proof of Work (PoW)' or 'Proof of Stake (PoS)' as consensus algorithms, but PoW and PoS are mechanisms to prevent sybil attacks, not consensus algorithms.

Since blockchain inherently includes a consensus function and additionally has the property of 'immutability,' meaning that once data is recorded, it cannot be modified or deleted, many believe that blockchain can easily create secure internet or mobile voting systems.

To conclude, this is incorrect. There are several conditions that a secure online voting system must satisfy, but the most important is 'software independence.' Software independence means that errors or unauthorized changes in the voting software should not affect the election results. To achieve this, voters must be able to directly verify whether their votes were accurately recorded as intended and reflected in the counting results. The problem is that achieving such software independence is very difficult in online voting systems using blockchain.

When using a public blockchain, where everything is transparently disclosed to everyone, the votes cast by voters are recorded on the blockchain and can be easily verified, making it easier to achieve software independence. However, since data recorded on a public blockchain can be seen by others besides the voters, guaranteeing secret ballots becomes difficult, which in turn creates a risk of vote-buying (paying voters to obtain votes). In other words, because one can directly verify whether a vote was cast for a particular candidate before paying, vote-buying becomes easier. To solve this problem, a private blockchain could be used, but this limits transparency and thus makes achieving software independence difficult. For these reasons, foreign experts and the National Academies of Sciences, Engineering, and Medicine (NASEM) have strongly criticized blockchain-based voting methods and are only considering limited application of blockchain to public voting methods such as Senate voting.

Currently, our government plans to develop a blockchain-based online voting system by 2022 and apply it to school principal elections, party primaries, and opinion-gathering votes. Elections are a much higher-risk area compared to virtual currencies. A more cautious approach from our government is required.

Seungjoo Kim, Professor, Graduate School of Information Security, Korea University

This content was produced with the assistance of AI translation services.