Fully Autonomous AI Could Control Criminal Networks: Europol

Europol has warned that blockchain and AI will “accelerate serious and organised crime” in the coming years, with a new report from the EU law enforcement agency even predicting that fully autonomous AI “could pave the way for entirely AI-controlled criminal networks.”

In its 2025 Serious and Organised Crime Threat Assessment, Europol suggested that the combination of AI and crypto will serve to expand online fraud schemes, which already represent “the most rapidly expanding sector” in organised crime.

“Narratives are extremely realistic, crafted with the help of AI, and incorporating trending societal topics,” the report’s authors wrote. “Cryptocurrency features prominently as a payment method.”

Crypto and crime

While not data-focused, the assessment offers a broad and far-reaching summary of the expanding role cryptocurrencies are now playing in serious crime in Europe.

The report concluded that cryptocurrency has “moved beyond the scope of cybercrime” and is found increasingly in more “traditional” areas such as “drug trafficking or migrant smuggling.”

Other varieties of crime which crypto facilitates include cyberattacks, online fraud, sanctions evasion, and also the trafficking of human beings.

Written by senior researchers in the fields of crime and security, the starkest warning comes with respect to how the emergence of AI may effectively help automate such criminal activity on an industrial scale.

SEC Launches New Crypto Crime Fighting Unit

The Securities and Exchange Commission is launching a new organization tasked with combatting crypto-related crime.  In a Thursday announcement, Wall Street’s top regulator said the new Cyber and Emerging Technologies Unit will work with the SEC’s crypto task force to “root out those seeking to misuse innovation to harm investors and diminish confidence in new technologies.” It will replace the Crypto Assets and Cyber Unit and will be made up of around 30 fraud specialists and lawyers from acros...

The report’s authors state that AI’s automation capabilities are already making criminal operations more efficient and expansive.

“From automating phishing campaigns to executing large-scale cyber-attacks, AI enables criminals to achieve more – reach more victims, be more targeted in their approach, and expand their global reach – with fewer resources,” the report read.

Among the cybercriminal uses of AI, the assessment lists “attack automation, social engineering, and bypassing security measures,” which again are making cyberattacks more scalable.

Europol’s evaluation of finds support in private research, with blockchain intelligence platform TRM Labs releasing a report in January of this year which revealed that bad actors “are increasingly leveraging AI to carry out hacks and conduct fraud.”

How criminals are using AI

Isabella Chase, the Head of Policy for EMEA at TRM Labs, told Decrypt that there are three primary ways in which criminals are already using AI.

The first relates to the automation of attacks, with AI-based tools enabling “the automation of phishing campaigns, creating highly convincing messages at scale,” according to Chase.

Related to this is how criminals are using AI to create deepfakes and other synthetic media.

“Criminals use AI to build convincing deepfakes or to embellish existing stolen identities to dupe victims,” she said. “This is increasingly important for business compromise, which is the most profitable form of online fraud.”

Hong Kong Deepfake Scam Group Caught Pretending to Be Rich Single Women

Hong Kong police have intercepted over HK$34 million (US$3.37 million) in scam proceeds as part of an elaborate operation targeting victims through the use of artificial intelligence. According to a report by the South China Morning Post on Saturday, the investigation also shed light on some of the novel strategies employed by local romance scam operators.  Some revolved around pretending to be wealthy single women, with conversational prompts including learning Japanese, playing golf, and tasti...

Less visibly, AI is also being used to support cybercrime and other exploits by providing criminals with additional information, providing data and details they may have otherwise missed.

As Chase explains, “AI algorithms optimize ransomware operations by identifying the most critical data or systems to encrypt for maximum leverage [...] AI also allows malware to adapt dynamically to evade real-time detection.”

The use of AI is likely already adding millions on to the cost of organised crime, but TRM Labs estimates that more criminals will adopt AI “in the coming years” as they experiment with different ways of harnessing it for their particular ends.

“Criminals will use AI for targeting and engaging with victims whether these be individuals or entities,” Chase told Decrypt, adding that the technology will also be used for “more programmatic laundering of criminal funds which could make it easier to profit from crime.”

 Artificial Intelligence

Chinese Police Roll Out Nightmarish Crime-Fighting Robot

In a scene right out of Black Mirror, police in China have rolled out the newest innovation in autonomous AI-powered street crime fighting: A spherical, virtually indestructible, amphibious, heavily armed device that resembles a giant bowling ball. The RT-G, manufactured by Shenzhen-based Logon Technology Co., Ltd., is the Swiss Army knife of crime-busting technology. According to reports, the robot is driven by AI, which can help it ID suspects via facial recognition and notify law enforcement...

While this may paint a bleak picture, Chase ultimately suggested that there will be a parallel growth in law enforcement and the private sector using AI and more advanced analytic tools, thereby improving “their ability to identify criminals and their activities.”