Neural network and scammers

How neural networks cheat: case studies, methods and protection

Artificial intelligence is being developed for the benefit of humans. AI tools can compose poetry, create works of art, generate videos and music, write texts, program and even joke. But while users are just learning about the capabilities of artificial intelligence, scammers are using neural networks to make illegal profits.

Although the uprising of machines predicted by science fiction writers has not yet happened, neural networks in the hands of attackers can go to the dark side. And they are so skillful that victims often do not even realize that they have been drawn into a fraudulent scheme until they lose all their savings.

Just imagine that someone close to you calls you and asks for help in a familiar voice with the same intonations. You transfer the amount to the specified card because “all of his accounts were blocked.” And only when the money has gone to the attackers, you realize that it was a deepfake. This is just one case of fraud using neural networks. In this article, we will analyze fraudulent schemes, methods of protection, and give recommendations for preventing attacks.

Typical schemes and examples of fraud

Every year, fraudsters become more advanced in digital technologies to find new ways to deceive citizens. This is a global problem that has become worse with the development of artificial intelligence.

Common schemes of cyber fraudsters:

• imitation of employees of financial institutions, law enforcement agencies;

• fake voices of relatives, colleagues, friends;

• creating fake videos with famous people or acquaintances in order to attract them to some actions on their behalf;

• fake online stores that look like real ones;

• extortion of money for fake personal photos or videos with explicit content;

• hacking accounts in social networks to send messages with requests for financial assistance;

• phishing links in emails generated by a neural network.

And this is only a small part of the types of fraud using AI. According to statistics from the Russian Ministry of Internal Affairs for last year, more than 20% of criminal cases opened under the article for fraud were related to crimes where neural networks were used.

Methods for detecting and preventing fraud

Despite the fact that criminal cases for fraud are opened regularly, it is quite difficult to find cybercriminals. They use digital technologies, proxy servers, encryption, intermediaries to transfer money from one account to another. Therefore, fraud with neural networks is easier to prevent than to punish criminals.

In Russia, strict measures are being taken at the state level to detect and prevent cyber fraud:

• New laws are being adopted to help identify the activities of fraudsters and the use of artificial intelligence in criminal schemes.

• Financial institutions, together with the Central Bank, monitor atypical transactions and block suspicious transfers.

• Since March 2025, a law has been in effect allowing citizens to issue a self-prohibition on issuing loans.

• Research is being conducted in the field of cybersecurity aimed at protecting personal data and transactions.

These measures help reduce the number of victims of fraudsters, increase the digital literacy of the population and the effectiveness of law enforcement agencies. However, everyone can protect themselves from fraudsters on their own, no matter how advanced the technology they use to commit crimes.

Recommendations

Fraudsters have become more inventive when they got their hands on artificial intelligence. But always remember that the human brain is smarter, and you can protect yourself.

Tips on how to protect yourself from cyber fraudsters:

1. Never trust information when asked to transfer money, give a code, passport details, home address. If a request for help comes from a loved one, check the information.

2. Do not transfer money under pressure. Fraudsters force the victim to act quickly.

3. Do not click on unfamiliar links, even if someone from your circle sent them to you.

4. If the caller’s voice seems suspicious to you or the person behaves strangely in the conversation, it is better to end the call and call back.

5. Use two-factor authentication to protect your accounts and create complex passwords.

In fact, AI, although unwittingly helping fraudsters, also participates in the fight against them. Developers are working on technologies that can detect deepfakes and block phishing attempts. Mobile operators use neural networks to analyze incoming calls and warn about spam. Some AI-based image generators do not agree to generate fake photos of branded products.

We hope that the information in the article will help you avoid unpleasant encounters with attackers using neural networks for fraud!