By 2025, 65% of cryptocurrency theft incidents involved social engineering attacks, with total annual fraud losses reaching $17 billion.

June 1st, Chainalysis research finds social engineering attacks have become the leading source of financial losses in the cryptocurrency industry. AMLBot data shows approximately 65% of cryptocurrency theft cases investigated in 2025 involved social engineering tactics, while Chainalysis estimates total global crypto fraud losses reached roughly $17 billion that year. Attackers prey on users’ trust, fear, and sense of urgency to carry out scams, rather than targeting the underlying blockchain technology directly. The report outlines six of the most common current attack methods: phishing, identity fraud scams, SIM swapping (hijacking), rug pulls (investment or romance scams), AI-driven scams, and airdrop and giveaway frauds. Hacken data indicates the Web3 sector suffered $306 million in losses from phishing attacks in the first quarter of 2026. Chainalysis data shows clusters of identity fraud scams saw a roughly 1,400% year-over-year increase. CoinLaw estimates SIM swapping attacks caused approximately $410 million in losses in 2025, while rug pulls and other investment scams accounted for around $7.2 billion in total losses—making this the largest fraud category overall. Furthermore, Chainalysis notes the average profit from a single AI-powered scam is about $3.2 million, roughly four times the average profit from traditional scam operations. U.S. FBI statistics show 22,364 complaints related to AI-assisted crypto scams were received in 2025, totaling roughly $893 million in losses. As attack methods continue to evolve, users should avoid relying solely on SMS verification codes as their primary security measure, verify customer service identities only through official channels, and regularly review and revoke wallet authorizations to reduce the risk of asset theft.