Best practices for detecting and preventing deepfake attacks


As public engagement for digital content continues to grow, consumers and businesses are increasingly reliant on technology platforms.

The anonymity of our digital world makes it difficult to see who is hiding behind the screen. This gray area offers potential fraudsters the opportunity to threaten both businesses and consumers directly, especially in the field of Deepfakes – Artificially created images, video and audio designed to mimic real human characteristics. In the past few years, Deepfakes have accumulated wide attention. They are a growing concern because of their involvement in fraudulent activities.

This is how AI deepfake technology works

Deepfake tactics allow fraudsters to distort reality by manipulating existing images to replace a person’s likeness. This tactic is based on artificial neural Networks – computer systems that recognize patterns in data. When developing a deepfake photo or video, hundreds of thousands of images are fed into the artificial neural network, which “trains” the data to identify and reconstruct facial patterns.

With the advent of more advanced AI, the number of images or videos required to train the artificial neural networks has decreased significantly, making it easier for scammers to use these tools on a large scale. Deepfake videos are widely used in financial crime to target individuals, businesses, and government regulators. The risks can be particularly acute in emerging markets or those with financial turmoil.

Best Practices for Detecting Deepfake Technology

Tools and best practices can help contain fraudsters’ efforts. The most important aspect is vigilance: scammers are relentless and always at work to exploit any loophole or weak point.

The first step is to look at the state of the deepfake videos for yourself. At this stage, it is often possible to spot deepfake video if you know what to look for. Some of the signs are as follows:

  • jerky movement;
  • Light shifts from one image to the next;
  • Changes in skin tone;
  • strange blinking or no blinking at all; and
  • poor lip-sync with the subject’s language.

Technologies are also emerging to help help video makers authenticate their videos. For example, a cryptographic algorithm can be used to insert hashes at set intervals during the video. If the video in question is changed, the hashes will change.

David Britton

Security procedures can go a long way in defending against fraudsters. As an emerging threat, deepfakes thrive at the level of technology available to scammers, in particular machine learning and advanced analytics. Firms can fight fires with fire by using the same defense skills.

A layered defense strategy is also vital, especially in relation to how scammers distribute or use deepfakes. The threat landscape is constantly evolvingSo there’s nothing more important than guarding the front door.

As risks and countermeasures evolve, the measures we are applying now are quickly becoming obsolete. As the nuances of deepfake technology continue to shift, key business best practices should stay the same. With awareness and vigilance, consumers and businesses can stay one step ahead of deepfake technology.

About the author

David Britton leads strategy and thought leadership for Experian’s Global Identity and Fraud Group. Britton has over 20 years of experience in digital identity and fraud. He brings a wealth of experience and unique insights into the criminal methodology behind cyber fraud, the evolving digital identity landscape, and the operational challenges of the business.


Leave A Reply