TLDR
- Interest in Blockchain technology and its applications has increased tremendously; however, this is accompanied by serious privacy threats.
- This paper studies users’ privacy perceptions of UTXO-based blockchains such as Bitcoin.
- It elaborates – based on interviews and questionnaires – on a mental model employing privacy-preserving techniques when doing blockchain transactions.
- It evaluates users’ awareness of blockchain privacy issues and examines their preferences towards existing privacy-enhancing solutions, i.e., add-on techniques to Bitcoin versus built-in techniques in privacy coins.
Core Research Question
- RQ1: To what extent are users aware of privacy issues and privacy-enhancing technologies?
- Do they trust third-party privacy-preserving services?
- To achieve better anonymity, do users prefer using add-on privacy techniques on top of current blockchain implementations (e.g., Bitcoin), or built-in features of privacy blockchains (e.g., Monero)?
- RQ2: Are users willing to use privacy-preserving techniques despite the higher fees and longer transaction time?
- RQ3: Which privacy features interest users the most (e.g., hiding the source, hiding destination, hiding the amount)?
Citation
Ghesmati, S., Fdhila, W., & Weippl, E. (2022). User-Perceived Privacy in Blockchain. Cryptology ePrint Archive. https://eprint.iacr.org/2022/287.pdf
Background
- De-anonymization in Bitcoin: All transactions on the blockchain are publicly available, anyone can apply specific heuristics in combination with auxiliary information (e.g., address tags) to cluster and identify users and their transactions. Thus, it becomes possible to link Bitcoin addresses to real identities.
-
Privacy-enhancing Techniques in Bitcoin: Add-on techniques on top of Bitcoin were proposed to enhance privacy. They can be categorized into four categories including centralized mixers, CoinJoin based, Atomic Swap, and Threshold signatures.
Privacy Coins: provide built-in privacy.- Monero: is a UTXO-based cryptocurrency that was developed to provide anonymity. It applies stealth addresses, ring signatures, and RingCT. Stealth addresses create a one-time address for each transaction. Ring signatures allow obfuscation of a sender by creating a group of inputs for hiding the actual sender.
- Zcash: is a Bitcoin fork that employs a type of zero-knowledge proof named ZK-SNARKS.
Summary
- We introduce the main concepts related to Bitcoin, privacy, and deanonymization attacks.
- We provide the result of our extensive research on privacy techniques.
- We investigate existing privacy wallets and top privacy coins.
- We describe our quantitative and qualitative studies.
- We present the results and discussion
Method
- Recruitment: Participants were recruited via social media, universities, and companies focusing on Blockchain technology.
- Interview Procedure: Before the interview, participants were briefly informed about the context of the research and signed a consent form. Each interview lasted about 30 minutes. We conducted semi-structured interviews both in-person and via online meetings.
- Coding: Grounded Theory was used for coding. The researchers coded the data and grouped statements related to the same concept.
- Sampling: We selected participants according to their reported level of knowledge and their usage of cryptocurrencies, ranging from expert to novice users. We interviewed 14 participants, 12 users (age: max. = 45, min. = 26) and 2 non-users (age: max. = 45, min. = 35). 7 out of the 12 users and 1 out of the 2 non-users were working in IT-related fields.
Results
-
Privacy Awareness:
- Some participants in the qualitative part did not know the difference between custodial wallets such as exchanges and non-custodial wallets.
- Some participants assumed that blockchain is safe from a privacy perspective as it uses addresses rather than real identities
- Users were willing to apply mitigation tools to enhance privacy in case of awareness.
- Awareness of address reuse and auxiliary information obtained from exchanges and services have increased among our participants. Most respondents were not aware of common input ownership, the most prominent heuristic in de-anonymization techniques
- Some participants had heard of Zcash and Monero, but were not aware they were privacy coins.
- Some of the participants assumed privacy persevering tools or privacy coins were used by criminals for tax evasion and refrained from applying privacy-enhancing measures or tools.
-
Privacy Preferences:
- While more than half of the participants preferred to use privacy coins rather than add-on techniques, most of those who chose to use add-on techniques on top of Bitcoin; expected future built-in privacy improvements to Bitcoin. Although this does not seem realistic in the near future, it is implemented by wallets or layer two solutions.
- Users are willing to accept longer transaction times to achieve better privacy, but half of them dismissed the idea of paying extra fees.
- Users who were aware of the distinguishability of CoinJoin transactions with the same output amount were not willing to use it. Instead, they favored alternative techniques that preserve indistinguishability, where the transactions cannot be flagged by monitoring tools.
-
Privacy Wallets:
- Although the development of bitcoin privacy wallets started around 2015, privacy wallets still struggle to attract users. These wallets are complex and require a minimum understanding of privacy concepts and techniques.
- On the one hand, current Bitcoin privacy wallets implemented CoinJoin with the same output amount suffer from distinguishability in the blockchain; on the other hand, the newly implemented indistinguishable techniques such as PayJoin or Wabisabi in Wasabi wallet 2 may be banned by governments, rendering them unusable for their respective wallet developers and users.
- Users prefer to use wallets that support different coins; thus, we can not expect users to install different wallets for different coins, let alone install additional wallets for their privacy, as well as having to spend time learning the wallet functions.
Discussion and Key Takeaways
-
Problems:
- Participants have little knowledge about privacy-enhancing techniques. Most of them were not informed about how these tools work, even if they had heard about the techniques or privacy coins’ names. They did not know what kind of privacy is provided by those techniques. or they assumed those techniques are too technical for ordinary users who only buy and sell cryptocurrencies for trading or investment purposes.
- There were also negative understandings of privacy tools (such as using them for criminal activities or tax evasion)
-
Solutions:
- The problems and consequences of the public availability of the data on the blockchain need to be educated through integration with wallets by providing meaningful notification and privacy features or through documentation and social media.
- We suggest the concept of public and private privacy, the former referring to information that is available through the blockchain to the whole world, and in the latter, the information that is available to governments, exchanges, or wallets.
Applicability
With a better understanding of users’ mental models, developers will be able to build more user-friendly privacy techniques or coins.