BYBIT’S 2025 SECURITY INITIATIVE SETS NEW BENCHMARK IN CRYPTO FRAUD PREVENTION

Cryptocurrency exchange Bybit has unveiled the results of its 2025 Security Initiative, introducing a multi-layered fraud prevention system designed to protect users before losses occur. The initiative intercepted hundreds of millions of dollars in suspicious withdrawals, signalling a shift towards proactive security across the digital asset industry.

As cryptocurrency adoption accelerates worldwide, the threat of digital fraud continues to grow in parallel. Against this backdrop, global crypto exchange Bybit has unveiled the results of its ambitious 2025 Security Initiative, presenting what it describes as an industry-leading, multi-layered defence architecture aimed at preventing fraud before financial losses occur.

The announcement comes at a time when scams in the digital asset sector remain a major concern. According to a recent report by blockchain analytics firm Chainalysis, approximately $17 billion worth of cryptocurrency was lost to scams and fraud schemes in 2025 alone. Such figures highlight the scale of the challenge faced by exchanges, regulators and users navigating an increasingly complex financial ecosystem.

Bybit, currently ranked among the world’s largest cryptocurrency exchanges by trading volume, says its new security framework marks a decisive shift from reactive protection to proactive risk prevention. Central to this strategy is a dynamic risk-based system that monitors user withdrawals in real time, identifying potential threats before funds leave the platform.

At the heart of the initiative lies what the company calls a Triple-Tier Withdrawal Fraudulent Defence Framework. Rather than applying a single blanket response to suspicious activity, the system categorises potential risks into three distinct levels, each triggering a tailored response designed to balance security with the smooth functioning of everyday trading.

The first level, described as the Early Warning tier, focuses on identifying low-risk anomalies through large-scale data analysis. Bybit’s monitoring tools analyse transaction patterns across the platform, searching for unusual behaviour such as multiple withdrawals directed towards a single newly created wallet address. When such patterns emerge, automated surveys are sent to the user, encouraging them to confirm the legitimacy of the transaction. The information gathered from these responses helps the exchange’s risk operations team determine whether a wallet address should be blacklisted to prevent further exposure.

The second layer, known as the Real-time Alert tier, is triggered when the system detects a higher degree of risk. Accounts may be flagged through credential-stuffing databases — collections of stolen usernames and passwords circulating across the internet — or when withdrawal requests are directed to suspicious addresses. When this occurs, users receive immediate warnings during the withdrawal process, prompting them to pause and review the transaction carefully. According to Bybit, the alerts are designed specifically to counter social engineering tactics, which often rely on creating a sense of urgency or emotional pressure to manipulate victims into transferring funds quickly.

The third and most stringent layer of protection comes into effect when the system identifies confirmed scam-related activity. In these high-risk situations, withdrawals to fraudulent wallet addresses are automatically blocked. The platform also introduces a mandatory one-hour cooling-off period, giving users time to reconsider the transaction and verify its legitimacy before funds can potentially be moved elsewhere. The measure is intended to interrupt sophisticated fraud schemes, including so-called “pig butchering” scams, where victims are gradually persuaded to invest increasing sums of money through deceptive online relationships or fake investment platforms.

According to the company, the impact of these measures has been significant. During the final quarter of 2025 alone, Bybit flagged approximately $500 million in suspicious withdrawal attempts. Of that amount, $300 million was successfully intercepted and recovered, preventing losses for more than 4,000 users.

Artificial intelligence played a central role in these outcomes. Bybit’s proprietary detection algorithms analysed on-chain blockchain data to identify fraudulent activity patterns. Through this system, the exchange detected 350 high-risk investment fraud wallet addresses, protecting around 8,000 users from potentially losing funds through compromised withdrawals.

The initiative also demonstrated resilience against large-scale hacking attempts targeting user accounts. Over the course of 2025, Bybit’s security infrastructure blocked more than three million credential-stuffing attacks, a common method used by cybercriminals to gain unauthorised access to accounts by testing stolen login information across multiple platforms.

In addition to automated monitoring, the company deployed both algorithmic and manual methods to track suspicious blockchain activity. During the fourth quarter, its system automatically labelled 350 wallet addresses as high-risk while risk control teams manually tagged a further 600 addresses following investigative review. These interventions alone helped prevent an estimated $1 million in imminent fraud losses.

However, Bybit emphasises that combating cryptocurrency crime requires cooperation beyond individual platforms. As part of its 2025 strategy, the exchange expanded its integration with external intelligence networks and blockchain monitoring organisations.

Among the partners contributing to its risk detection system are the blockchain analytics firms TRM Labs, Elliptic and Chainalysis. By combining on-chain monitoring technology with shared industry intelligence, the exchange aims to map patterns within fraudulent networks and identify suspicious activity earlier.

David Zong, Head of Group Risk Control at Bybit, said the company’s ambition is to transform the traditional role of security systems within cryptocurrency platforms. Rather than operating as a passive safeguard in the background, he explained, the goal is to create an active protection mechanism that constantly evaluates and responds to risk in real time.

“Our mission in 2025 was to transform risk control from a ‘silent shield’ into an active, intelligent guardian,” Zong said. By combining artificial intelligence-driven monitoring with shared intelligence from industry partners, he noted, the exchange hopes not only to protect its own users but also contribute to broader efforts aimed at dismantling organised fraud networks.

The collaborative approach reflects a growing recognition across the cryptocurrency sector that cybercrime cannot be tackled by isolated efforts. Fraud operations frequently span multiple exchanges, jurisdictions and blockchain networks, making collective intelligence sharing a crucial tool for identifying patterns and tracking illicit flows of digital assets.

Bybit says it has begun sharing standardised monitoring indicators with partners across the ecosystem to improve detection capabilities industry-wide. The exchange believes that strengthening transparency and cooperation between platforms, analytics firms and regulators will ultimately create a safer environment for both experienced traders and new entrants to the crypto market.

As digital assets continue to move closer to mainstream financial adoption, the pressure on exchanges to protect users has never been greater. While technological innovation has opened new opportunities for global finance, it has also created sophisticated avenues for exploitation by cybercriminals.

The results of Bybit’s 2025 Security Initiative suggest that proactive, data-driven protection systems may become an essential component of the next generation of cryptocurrency platforms. Whether other exchanges adopt similar frameworks could play a significant role in shaping the future security landscape of the digital asset economy.