Typosquatting 3.0: Characterizing Squatting in Blockchain Naming Systems

Muhammad Muzammil, Zhengyu Wu, Lalith Harisha, Brian Kondracki, Nick Nikiforakis

This research paper, titled "Typosquatting 3.0: Characterizing Squatting in Blockchain Naming Systems," provides the first large-scale analysis of how attackers exploit human error within decentralized naming services like ENS.

TL;DR

In the world of Web3, a single typo doesn't just lead you to a fake website—it results in the immediate and irreversible loss of your funds. This paper characterizes "Typosquatting 3.0," revealing that attackers are aggressively registering variations of popular blockchain names to intercept thousands of dollars in accidental transactions, while wallet providers offer almost zero protection.

What Problem Does This Paper Address?

Blockchain Naming Systems (BNS)—such as the Ethereum Name Service (ENS)—were designed to make crypto more user-friendly by replacing long, hexadecimal wallet addresses with human-readable names like alice.eth.

However, the authors identify a critical security gap: intra-BNS typosquatting. In traditional DNS, a typo might lead to a phishing site where an attacker still has to trick you into giving up data. In BNS, the name is the destination for a financial transaction. If you mistype a recipient's name, the blockchain protocol faithfully executes the transfer to the squatter's wallet. There is no "undo" button, no middleman to dispute the charge, and, until now, very little data on how prevalent this threat actually is.

Methods

The researchers developed a multi-stage pipeline to quantify this ecosystem:

  1. Data Harvesting: They used decentralized indexing protocols (like The Graph) and blockchain explorers (Etherscan, Polygonscan, Blockfrost) to bypass the "namehash" obfuscation used by BNS, allowing them to see the actual plaintext names.
  2. Establishing Ground Truth: Since there is no "Top 1M" list for BNS names, they created a popularity metric based on transaction volume ($w_T$) divided by the number of domains pointing to a specific wallet ($w_D$). This helped distinguish truly popular targets from "squatter nests."
  3. Typo Generation: They applied seven classic typosquatting models (addition, omission, swapping, etc.) to popular names to identify registered squatting candidates.
  4. Victim Identification: To prove actual harm, they looked for "common senders"—wallets that had sent funds to both a legitimate name and its typo-variant. This provided a conservative lower bound for successful attacks.

Datasets

The study utilizes the largest BNS dataset compiled to date:

  • 4.9 Million BNS names across three services: Ethereum Name Service (ENS), Unstoppable Domains (UD), and ADA Handles (ADAH).
  • 200 Million transactions analyzed to track fund flow.
  • Social Media Corpus: A curated list of BNS names from 14,910 Twitter/X profiles to study how social media popularity correlates with being targeted.

Key Insights and Ideas

  • The "Surround" Attack: Unlike traditional squatting where an attacker might grab one or two domains, BNS squatters often "surround" a target. For example, vitalik.eth (belonging to Ethereum's creator) has at least 74 registered typo variations resolving to 66 different attacker wallets.
  • Social Media as a Target Signal: Attackers aren't just guessing; they monitor Twitter/X for "influencers" who advertise their BNS names. The researchers found that 38% of popular ENS/UD names found on Twitter were targeted by at least one squatter.
  • The Defensive Gap: In the traditional web, 31% of top brands proactively register typos of their domains to protect users. In BNS, that number drops to a staggering 0.1%. Most users simply don't realize they need to "defensively register" variations of their own names.
  • Industrial-Scale Squatting: The researchers identified three distinct squatter behaviors: those who target a single name, those who target many dissimilar popular names, and "power squatters" who hold thousands of domains in a single wallet.

Key Results

  • Prevalence: Up to 37% of the most popular ENS domains have at least one registered typosquatter.
  • Financial Impact: The average successful scam transaction sent $1,790 to a squatter. In some cases, victims sent funds to squatters multiple times, likely because they saved the "typo" address in their transaction history.
  • Growth Trends: Typosquatting registrations have exploded since 2021, closely correlating with the rise in cryptocurrency prices.
  • Wallet Failure: The authors tested 10 popular wallets (including MetaMask and Coinbase). None of the major wallets provided warnings for "cold typos" (sending to a squatter for the first time). Only two niche wallets (Eternl and Bitcoin.com) provided a generic "transactions are irreversible" warning after name resolution.

Limitations and Future Work

The authors acknowledge that their "common sender" approach is a lower bound. It misses victims who made a typo on their very first transaction to a recipient. Furthermore, "aliasing" remains a challenge: it is difficult to determine if a user typed a name or pasted a direct wallet address from on-chain data alone.

Future Work: The paper calls for "Typosquatting 3.0" defenses to be integrated directly into wallets. This includes local fuzzy-matching (warning a user if vitalic.eth looks suspiciously like a previously used vitalik.eth) and the use of global, community-maintained blocklists of known squatter addresses.

Summary updated

Abstract

A Blockchain Name System (BNS) simplifies the process of sending cryptocurrencies by replacing complex cryptographic recipient addresses with human-readable names, making the transactions more convenient. Unfortunately, these names can be susceptible to typosquatting attacks, where attackers can take advantage of user typos by registering typographically similar BNS names. Unsuspecting users may accidentally mistype or misinterpret the intended name, resulting in an irreversible transfer of funds to an attacker's address instead of the intended recipient. In this work, we present the first large-scale, intra-BNS typosquatting study. To understand the prevalence of typosquatting within BNSs, we study three different services (Ethereum Name Service, Unstoppable Domains, and ADAHandles) spanning three blockchains (Ethereum, Polygon, and Cardano), collecting a total of 4.9M BNS names and 200M transactions-the largest dataset for BNSs to date. We describe the challenges involved in conducting name-squatting studies on these alternative naming systems, and then perform an in-depth quantitative analysis of our dataset. We find that typosquatters are indeed active on BNSs, registering more malicious domains with each passing year. Our analysis reveals that users have sent thousands of transactions to squatters and that squatters target both globally popular BNS domain names as well as the domains owned by popular Twitter/X users. Lastly, we document the complete lack of defenses against typosquatting in custodial and non-custodial wallets and propose straightforward countermeasures that can protect users without relying on thirdnartv services.
BibTeX
@inproceedings{muzammil2024typosquatting,
  author = {Muhammad Muzammil and Zhengyu Wu and Lalith Harisha and Brian Kondracki and Nick Nikiforakis},
  title = {{Typosquatting 3.0: Characterizing Squatting in Blockchain Naming Systems}},
  booktitle = {eCrime},
  year = {2024},
  url = {https://openreview.net/forum?id=XKT2dzPrGe},
}

Loading related content...

Figure