The development of modern technology has brought many innovations, and one of the new technologies that is shaking up the media industry is deepfake. DeepFake is a video, image or sound recording that has been processed using AI technology. In a word, it will turn the person in the video into another person by combining faces, voices, mouth shapes, etc.
Deepfake content is highly persuasive, and the constant development of deepfake technology makes it more difficult to distinguish between real and fake content. While deepfake technology is still relatively new, we are still seeing its role in phishing and cybercrime trends. This has become a growing concern for consumers and organizations alike, as phishing criminals take advantage of deepfakes to carry out social engineering attacks and spread misinformation.
The losses of deepfake scams are estimated to exceed $250 million in 2020, and the technology is still in its early stages. No doubt as deepfake technology evolves, so does the sophistication of criminals exploiting the technology.
What is Deepfake?
Deepfake is a form of media that stitches an existing image or video with AI-generated content to resemble someone’s voice or appearance. Often referred to as a form of “synthetic media,” DeepFake mimics people’s faces, movements, and voices with such precision that it’s impossible to tell the real thing.
Thanks to sophisticated machine learning algorithms, biometrics such as facial expressions and pitch of human voices can be manipulated to create expressions and actions about events that never occurred. While not all deepfakes are used with malicious intent, this form of digital impersonation is often used to create fake videos and audio recordings of people doing or saying malicious things. harmful.
How does Deepfake technology work?
The most important factor to create deepfake technology is Machine Learning. DeepFake is based on an AI computing system known as an artificial neural network, which is based on the human brain and is designed to recognize data patterns. Here’s how to start creating deepfakes.
To create a deepfake video, the creator starts by feeding hundreds of hours of real video to an artificial neural network to “train” the computer to identify detailed human features. It is intended to give the algorithm realistic information about what the person looks like from different angles.
The next step involves combining a trained neural network with computer graphics techniques to overlay real footage of other people and AI-synthesized voices.
While many people believe that creating deepfake videos requires complex tools and specialized skills, that’s not true — they can be created even with basic computer graphics skills. More convincing videos require more advanced techniques, but all you really need is access to someone’s video or audio footage. This is very easy to do with today’s huge amount of data, resulting in an abundance of material to feed the algorithm and create an actual deepfake video.
3 types of risks from Deepfake
The rapid advancement of deepfake technology has created opportunities for technology criminals to cause serious financial harm. From identity theft and public disinformation to corporate extortion, phishing and automated cyberattacks, deepfake technology has created a new kind of media that bad actors are using. to redeem. Here are some ways criminals use deepfakes.
Ghost scam
Ghost fraud occurs when a criminal steals the data of a deceased person and impersonates them for financial gain. Stolen identities can be used to access services and online accounts, or to apply for things like credit cards and loans.
New account scam
Also known as application fraud, new account fraud involves using a stolen or forged identity for the purpose of opening a new bank account. Once criminals have opened an account, they can cause serious financial damage by using credit cards or borrowing money with no intention of paying it back.
Synthetic identity fraud
Synthetic identity fraud is a more sophisticated method of fraud and is often harder to detect. Instead of exploiting one person’s stolen identity, criminals exploit the information and identities of many people to create a “person” that doesn’t actually exist. This identity is then used for large transactions or new credit applications.
Examples of Deepfake Attacks
Recent new advances have dramatically reduced the amount of time and data required to create highly accurate DeepFakes. As DeepFake becomes more accessible, the number of attacks will also continue to grow. The examples below will give you a glimpse of the capabilities of deepfake technology.
Energy company scam
This deepfake scam from 2019 is the first deepfake attack and it shows everyone the dark side of deepfake technology. The CEO of an energy company received a phone call from what he thought was the company’s boss and chief executive officer. In fact, that voice was actually an AI-generated voice, and that’s why he quickly agreed to an urgent request to transfer $243,000 within an hour.
Tech company scam
In a failed attempt to test deepfake audio from 2020, an employee of the tech company received a strange voicemail from someone who appeared to be the company’s CEO. Asked for “immediate assistance to finalize an urgent business deal” and the employee became suspicious and took it to the company legal department.
While this deepfake attack ultimately fails, it’s a typical attack that we’ll see more of as technological advancements become more prevalent.
DeepFake Scam has appeared in Vietnam
I have just read a few articles recently, in Vietnam there is an object that has scammed artists to make Facebook green ticks, thereby taking over the accounts and defrauding the relatives of that artist. “In particular, this criminal group uses technology to fake the image of the Facebook owner in calls with the victim, and at the same time opens a bank account with the same name as the facebook account owner to build trust with the victim. individuals and propose to transfer money to these accounts to appropriate.”
How to spot a Deepfake scam
Although the scammer in a deepfake attack often masquerades as a director, the deepfake’s gentle tone makes employees suspicious. Some similar signs that can help you recognize it as a deepfake industry include:
- The rhythm of speech is not natural
- Robot-like melodies
- Blinking or unnatural movements
- Lip movements out of sync with speech
- Poor audio or video quality
Organizations can stay safe from a deepfake attack mainly by training their employees to identify through audio or through tests with the adversary. Often, training employees in cybersecurity measures is an organization’s first line of defense against cyberattacks.
The future of DeepFake
Deepfake technology has seen incredible progress in just a few short years. While deepfake detection tools are improving, so is the deepfake itself. As a result, recent efforts have focused on further research to understand deepfake technologies and more defenses.
While deepfakes are certainly not the first threat to cybersecurity, they represent a growing challenge that requires ongoing research to prevent criminals from exploiting it. Organizations and individuals will also need to find new ways to properly secure their data against increasingly sophisticated cyberattacks.