The risks of Artificial Intelligence: Deepfakes
Over the last few years, we’ve experienced the rapid evolution of Artificial Intelligence (AI) and accepted it as part of our day-to-day. However, despite its benefits, AI introduces new challenges that are hard to navigate. In this article, we explore the growing concerns around deepfakes and cyber threat.
What are deepfakes?
The term ‘deepfake’ is defined as media that uses AI algorithms to create images, videos, or audio recordings that look real – but aren’t. While these can be useful tools in some areas of entertainment and education, deepfakes are often misused to spread misinformation, impersonate individuals, and manipulate public opinion.
Real-life examples
- In 2023, a cybercriminal used a voice deepfake to impersonate an IT staff member, convincing an employee to divulge a multi-factor authentication code, granting unauthorised access to company systems.
- In another case, scammers used an AI-generated video to mimic a company’s CFO during a video call, deceiving a finance worker into transferring $25 million into their own accounts.
How deepfakes target businesses and individuals
As well as financial fraud, deepfakes also have the potential to cause serious damage to the person or businesses’ reputation who is being imitated. As AI technology becomes more accessible, the likelihood of this type of activities goes up, opening up the possibility of using the technology for revenge, blackmail, fraud and so on.
How to identify a deep fake
The first thing to consider when it comes to identifying a deepfake is the context of the situation. Is the person acting how you would usually expect or are they saying or doing things that seem out of character? There can be inconsistencies with deepfakes, but some can look and sound incredibly convincing, so ensure to question anything you see or hear that’s ‘not quite right’.
You may also notice:
- Unnatural facial movements or expressions
- Inconsistent lighting or shadows
- Audio and mouth not syncing
- Distorted backgrounds or objects
Protecting yourself and your business against AI-threat
Employee training: Educate staff about the risks of deepfakes and the importance of verifying unusual requests, especially those involving financial transactions or sensitive information.
Use multi-step verification: to better protect your business and devices.
Cyber insurance: Consider policies that offer protection against cyber threats, providing financial support in the event of an incident.