AI fuels 440% surge in hackers using blockchains in attacks

Open-source artificial intelligence has caused a 440 per cent rise in hackers embedding malicious code in blockchains, according to Chainalysis. According to the report, the practice of embedding malicious code in blockchains makes attacks harder to detect and block than conventional methods. The finding is attributed to Chainalysis, a blockchain-analysis firm.

The report states that state-backed groups from countries such as North Korea and Iran lead most blockchain-based malware activities. It says such cases averaged two per day before the release, in the middle of 2025, of powerful Chinese open-source AI models. Those models, per the report, carry no restrictions on generating malicious code. The excerpt does not name the specific models, the affected blockchains, or a total count of incidents.

Chainalysis attributes the 440 per cent increase to the availability of those models. The report says the release of the models in mid-2025 was followed by the surge in blockchain-based malware embedding, and that state-linked actors account for the majority of the activity. The excerpt does not state which organisations detected the increase, what defensive measures the report recommends, or whether any jurisdiction has responded with legislation or enforcement action.