Main menu


Why deepfake technology is a growing cybersecurity concern

featured image

According to Cyber ​​Magazine figures, hacker interest in using deepfake technology has surged 43% since 2019. Recent incidents related to the use of deepfake technology also demonstrate the increasing use of the technology for malicious purposes. .

In this interview, Kumar Vaibhav, Solutions Architect at In2IT Technologies, discusses what deepfake technology is, why its use cases are on the rise, and how governments and corporate organizations can protect themselves from this growing cybersecurity threat. talking about how.

Can you tell us more about the evolution of deepfake technology over the years?

In layman’s terms, it is a technology that processes the subject’s face into the subject’s face and the subject’s voice. It can also be the other way around, where the voice is changed to the target’s face.

In more scientific terms, the technology relies on artificial neural networks. Let’s say you have an iPhone and need to log in with facial recognition. The iPhone is biometric or captures every node on your face and that’s how it authenticates you.

Deepfake technology works similarly. A computer system that recognizes patterns in data. Creating deep fake photos and videos requires feeding hundreds of thousands of images into a trained artificial neural network. The algorithm then trains the data to identify and reconstruct facial or vocal patterns.

It can be used for legitimate purposes like deepfakes where David Beckham spoke nine languages ​​to raise awareness about malaria. It may be used for some purpose.

How hot is deepfake technology in the world right now?

Some serious incidents have occurred recently, one of the incidents related to the war between Russia and Ukraine. What happened is that around March, a video was posted on social media depicting an image of the Ukrainian president instructing his soldiers to surrender to Russian forces.

Another recent incident was when the cryptocurrency project team was tricked into believing they were meeting with Binance executives to allow their tokens to be listed on financial platforms. We took a public video of a specific executive and digitally processed that video to create an AI hologram.

What are the main reasons for the increase in deepfake technology attacks over the last few years?

It’s not a one-step approach when a cyberattack occurs. This includes reconnaissance procedures. It’s about understanding how to infiltrate an environment, steal data, and restore traces so no one can find you.

One of the reasons deepfake attacks are increasing year by year is that the AI ​​technology used to create deepfake video and audio far outpaces the pace at which companies are training their people to be able to deal with attacks. It’s evolving at a fast pace. .

For example, by the time companies figure out how to stop voice deepfake attacks, attackers have already created video deepfakes. This means that companies are always one step behind.

Another factor is probably the lack of adequate cybersecurity budgets. To keep up with technology evolving as rapidly as AI requires funding, most companies are unable or unwilling to invest in the necessary cybersecurity infrastructure and personnel.

Going back to South Africa, how prominent is the use of deepfake technology in this country?

I haven’t seen much in South Africa yet. But as this technology evolves rapidly and everything is interconnected on the Internet, it could easily escalate from country to country or continent to continent.

All it takes is for the bad character to identify the opportunity and take advantage of it. Major events such as conflicts and elections are great opportunities for these characters to use deepfake media. It’s just a matter of time.

How can South African companies avoid falling victim to deepfake attacks?

Microsoft has something called a video authentication tool that isn’t 100% effective, but I think it’s about 70% effective. It’s a start. Just put a video in there and it will tell you if the video is real or not.

There are no 100% effective tools on the market to combat deepfake attacks, so the best approach today is to understand what they are and how they work. Vigilance is another important factor in preventing attacks.

How can the impact of such attacks be mitigated if individuals or businesses are affected?

Awareness is key, as it is impossible to hide manipulated videos if they are already available to the public.

Education about deepfake technology involves educating the public on how to distinguish between genuine content and digitally altered content.

Are there countermeasures that can prevent deepfake attacks?

Apart from the aforementioned Microsoft tools, there are also developments to flag deepfake content using cryptography.

You can encode some kind of encryption algorithm into the original video so that it will alert you if someone tries to manipulate it.

But deepfakes are evolving rapidly. Do you think these counter technologies can catch up?

It’s difficult, but not impossible. Tackling this problem requires more investment in cybersecurity by organizations and governments, and that investment is currently lacking.

The potential for deepfake technology to cause widespread damage should motivate relevant stakeholders to accelerate efforts to prevent it.

Do you think there is a role that regulation can play in reducing the prominence of deepfake attacks in South Africa?

South Africa has the POPI law, which prevents anyone from disclosing another person’s personal information. But in a deepfake attack, the attackers are trying to trick you, so this regulation misses the point.

In the United States, there are laws that say if someone creates deepfake content for malicious purposes, they can go to jail. So South Africa can also follow this route.

But just as campaigns have gone beyond the law to educate the public and tackle misinformation about COVID-19 and vaccines, so too must awareness of technology and its implications.

11. Please share your farewell thoughts with our readers on deepfake technology.

Deepfake technology is still relatively new and evolving rapidly, so the best way to tackle its malicious use cases is education, education, education. You need to be proactive rather than reactive.

Deepfake content can have far-reaching consequences if not dealt with properly, as seen in cases like the incident involving the president of Ukraine, and is therefore combined with Microsoft tools and technologies such as cryptographic hashes. and is best dealt with with awareness tactics.

On Friday 23rd September, TechCabal will partner with Moniepoint (by TeamApt) to host the most important players in technology and business from across the continent and beyond to discuss the future of commerce in Africa. Register now to participate.

Get the best African tech newsletters in your inbox