Artificial intelligence and crypto scams: How digital assets are stolen today

Traders Union
All News
Editors' Picks
Artificial intelligence and crypto scams

All news

Editors' Picks

Crypto News

Financial News

Market Voices

Billionaires

Companies

Broker News

Regulators News

Olesia Kramarenko

15.01.2026

Artificial intelligence and crypto scams: How digital assets are stolen today — Artificial intelligence and crypto scammers: The new mechanics of digital asset theft

In 2025, crypto scammers reached a new level. Impersonation, social engineering, and artificial intelligence ceased to be separate tools and merged into a single, well-coordinated mechanism for stealing cryptocurrencies. What only a few years ago looked like primitive phishing with broken English now resembles a full-fledged industry built on automation, personalization, and a precise understanding of human psychology.

As a result, security concerns for crypto investors are becoming increasingly critical. The issue is no longer about technical hacks but about how artificial intelligence is reshaping the very nature of deception and why even experienced users are becoming victims more often.

The scale of the problem: what the numbers show

According to Chainalysis, 2025 became a record year in terms of losses from crypto fraud. Analysts estimate that at least $14 billion in cryptocurrencies was stolen through scam schemes, and this figure is not final. Considering hidden addresses and transactions that become visible with delay, total losses could reach $17 billion.

These figures matter not only on their own. They point to a shift in the nature of criminal activity. Crypto fraud is no longer a peripheral phenomenon associated with naive newcomers. It has become a systemic source of income, comparable in scale to entire segments of traditional financial crime.

Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

Impersonation as the primary attack vector

So-called impersonation schemes grew fastest in 2025. In these cases, scammers pose as trusted entities such as exchange support teams, well-known crypto projects, influencers, or even government institutions. Chainalysis recorded growth of more than 1,400 percent compared to the previous year.

Even more telling is another metric. The average payment in these schemes increased more than sixfold. This indicates that scammers have learned not only to send messages at scale but to work methodically with victims, gradually building trust and ultimately pushing them toward transferring significant sums.

The role of artificial intelligence

This is where artificial intelligence becomes the key factor. According to Chainalysis estimates, scam schemes that use AI tools generate, on average, 4.5 times more revenue than traditional approaches. The reason lies not in the “intelligence” of the models but in their ability to scale social engineering.

AI makes it possible to produce error-free texts that are stylistically precise and tailored to a specific individual. It can conduct real-time conversations, adjust arguments based on a counterpart’s reactions, and maintain contact for weeks. Some schemes also use voice cloning or video deepfakes that imitate real people, further blurring the line between fraud and legitimate communication.

Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

It is also cheap. What once required a team of operators can now be handled by an automated system capable of interacting with hundreds of potential victims simultaneously.

Social engineering instead of technical hacks

Despite the sophistication of the tools, the essence of most attacks remains unchanged. Blockchains and smart contracts are not being hacked. People are persuaded to hand over access themselves or to sign transactions voluntarily.

This is particularly effective in the crypto environment because transactions are irreversible. If a user willingly signs a transfer or interacts with a malicious contract, no bank and no arbitrator can roll the operation back. Artificial intelligence does not introduce new vulnerabilities here, but it exploits old ones with maximum efficiency: fear, urgency, authority, and trust.

How risks are changing for crypto investors

As a result, the risk profile for investors has changed significantly. Threats no longer look like an obvious phishing email or a suspicious website. They arrive in the form of a convincing dialogue, a familiar brand, and a logically constructed scenario. Even experienced users who understand the technical side of cryptocurrencies well can make mistakes in situations where pressure is applied gradually and psychologically.

This means that classic security advice no longer works as a universal safeguard. It remains necessary but insufficient in an environment where deception looks almost identical to legitimate communication.

How law enforcement and the industry are responding

Law enforcement agencies and analytical firms are trying to adapt to the new reality. They are making greater use of blockchain analysis, sharing data across jurisdictions, and cooperating with exchanges to block known scam-related addresses. At the same time, security services are also beginning to apply AI to detect anomalies and suspicious behavior.

Still, this struggle resembles an arms race. The same technologies used for defense are also available to scammers. And for now, the speed at which attackers adapt often outpaces the response of regulators.

For investors, this leads to a simple but uncomfortable conclusion. The main risk today lies not in the code, but in communication. And in a world where deception is becoming increasingly professional, the most valuable security tool remains the ability to slow down, question, and verify even what appears completely legitimate.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.