The AI Crime Gap: Why Criminals are Always Ahead of the Police

Criminals now use AI every day to clone voices, write flawless phishing emails, and scale scams globally. Meanwhile, some jurisdictions ban their own investigators from touching AI tools, a Recoveris executive warns.

The warning came during BeInCrypto’s Market Intelligence Council panel on crypto crime. Sol Cinosi of Recoveris and Nick Pailthorpe of Kodex agreed on the diagnosis. The AI crypto crime fight is becoming an arms race, and one side shows up unarmed.

Grammar Checks Don’t Catch Phishing Anymore

Cinosi, a former Buenos Aires prosecutor who now leads government and corporate affairs at Recoveris, said during the experts panel that AI is changing the game for both sides. On the criminal side, it scales sophistication and reach at once.

“The old advice to look for poor grammar in phishing emails doesn’t really work anymore. AI clones voices, generates deepfakes, and automates scams at a scale that simply wasn’t possible a few years ago.”

The numbers back him up. Crypto scams took an estimated $17 billion in 2025, according to the Chainalysis 2026 Crypto Crime Report.

Scam operations with on-chain links to AI vendors extracted $3.2 million each, 4.5 times as much as those without. AI does not just make scams cheaper. It makes them dramatically more profitable.

Some Police Are Forbidden From Using AI

The other side of the race looks very different. Cinosi named it as his single biggest worry.

“My biggest concern right now is law enforcement, because many agencies are still behind in AI adoption. Some don’t yet have clear policies, and in certain jurisdictions, investigators are forbidden from using AI tools.”

The result is an asymmetry that regulators are only starting to acknowledge. Europol’s IOCTA 2026 report calls it a “velocity gap” and says enforcement now depends on bridging it with technology. Cinosi put it more bluntly.

“If criminals are using AI every day and investigators, particularly law enforcement, aren’t, we are creating an uneven playing field. That’s not a race we’re going to win.”

Notably, he argued the barrier is often softer than a formal ban. Many investigators simply fear the tools, assuming they lack permission to use powers they already have. In his view, the gap is as much about capacity building as regulation.

AI as the Investigator’s Force Multiplier

The irony is that tracing technology has never been better. Cinosi said Recoveris now follows funds across blockchains, bridges, and even mixers with high confidence. Technology rarely stops an investigation anymore. People do.

“For investigators, AI helps solve a capacity problem by processing enormous amounts of data faster and identifying patterns and connections that would take much more time to find manually.”

That capacity matters because crypto adoption is growing faster than the number of experts. Experienced blockchain investigators remain scarce, so every hour AI saves goes where machines still fail.

“It allows us to spend more time doing what humans still do best, which is applying judgment and making critical decisions.”

AI Crypto Crime: The Training Gap on the Front Line

Pailthorpe, who spent 20 years in UK policing before joining Kodex, sees the same gap from the street level. Crypto no longer appears only in crypto cases. He meets it daily in counterterrorism and human trafficking investigations.

“What I’m seeing is more and more investigators going out to more and more scenes and realizing that crypto is part of that investigation, and not necessarily knowing what to do when they come across a wallet.”

The industry response is constant training. Pailthorpe runs webinars and police station visits almost daily, while exchanges on Kodex provide educational material to verified officers. Still, he refused to end on alarm.

“There’s that challenge of the criminals being one step ahead, but I am confident in lots of police agencies around the world.”

The tools exist, and the trainers are working. Whether policy lets investigators use the same weapons as their adversaries may decide who wins the race.

The post The AI Crime Gap: Why Criminals are Always Ahead of the Police appeared first on BeInCrypto.