.png){kind=logo}
Trugard and Webacy have launched a new AI system that prevents over 97% of attacks aimed at crypto wallet addresses. At the time of its release, the cryptocurrency sector was experiencing an increase in cybersecurity issues, with address poisoning attacks leading to millions of dollars in losses for users. This machine learning platform represents a significant advancement in enhancing the security of blockchain for individuals who hold digital assets.
Trugard and Webacy have collaborated to introduce an AI-based tool designed to tackle crypto address poisoning. The companies announced the release on May 21, highlighting that the system forms part of Webacy’s broader suite of crypto decision-making tools. The solution leverages a supervised machine learning model trained on live blockchain transaction data. It also incorporates advanced on-chain analytics, feature engineering, and behavioral context to improve accuracy.
According to Webacy co-founder Maika Isogawa, many overlook address poisoning as a devastating crypto scam. Attackers fool users by using addresses similar to those of the intended recipient. The method takes advantage of people who copy or use addresses from the clipboard, causing them to send funds to scammers by accident. Research shows that over 270 million addresses attempted to poison BNB Chain and Ethereum from July 2022 to June 2024, resulting in over $83 million lost from successful scams.
Jeremiah O’Connor, the company’s chief technology officer, highlighted that they possess a broad cybersecurity background, mainly from the Web2 industry. He stated that traditional crypto security tools often lag behind sophisticated attacker methods, relying mostly on static rules and simple filters. In contrast, their new system adapts by learning from new attack patterns, ensuring it stays ahead of evolving threats.
The AI tool’s machine learning model was carefully trained using real and test transaction data. The model employs supervised learning, where it studies labeled transaction behaviors to guess the chances of an attack. Given real-time feedback from artificial data, Trugard continuously tests and updates its model to react to new threats.
O’Connor emphasized the importance of context and pattern recognition in their approach. Instead of filtering transactions by static rules, the AI evaluates behaviors and evolving trends. This methodology has resulted in a reported 97% detection rate, significantly improving protection for users across decentralized finance and crypto wallets.
The ongoing training and adaptation of the model mean that it improves over time. By integrating new data as fresh tactics appear, the system can respond faster to changes in scam techniques. This feature offers a layer of defense not commonly found in traditional anti-fraud tools within the Web3 ecosystem.
Users on leading blockchain networks are still at risk from poisoning attacks. Criminals steal cryptocurrency by mimicking wallet addresses, taking advantage of the fact that the letters and numbers in hexadecimal look similar. As more people use cryptocurrencies, these scams are becoming more common, putting billions of assets on public networks at risk.
Trugard and Webacy secure crypto users with AI, machine learning, and advanced protection. The tool is designed to address ongoing problems and quickly respond to hackers' new tricks. As blockchain progresses, AI-powered security will be adopted as the norm to protect transfers and finances.
Also Read: Bitcoin Wallet Security: How to Prevent Fraud and Protect Your Crypto Assets
