Deepfake cybercrime is escalating as artificial intelligence tools become more accessible and sophisticated, according to cybersecurity experts. The technology enables criminals to create convincing fake videos and images for fraud, identity theft, and corporate scams.
Recent investigations into social media platforms highlight the growing threat. Last week, authorities examined Musk’s platform X following reports of sexualized deepfake images generated by its Grok chatbot. The European Commission and several countries launched inquiries into the matter.
How Deepfakes Enable Criminal Activity
The ease of creating deepfakes was demonstrated by a viral video showing Hollywood actors Tom Cruise and Brad Pitt in a fabricated fight scene. Filmmaker Ruairi Robinson stated the video was generated from a two-line prompt, illustrating how simple the process has become.
Konstantin Levinzon, co-founder of Planet VPN, said the rise in AI-driven technologies has made it easier for malicious actors to deceive people on an unprecedented scale. The company provides virtual private network services with a no-logs policy.
“A particularly dangerous aspect of AI technologies is that they make it difficult to distinguish what’s real. The internet is flooded with fake images and videos, also known as deepfakes, which can be created in just seconds with low-cost or even free tools,”
Konstantin Levinzon, Co-founder, Planet VPN
Corporate Targets and Financial Fraud
Businesses face significant risks from cybersecurity threats involving deepfakes. Criminals target employees with videos impersonating chief executives to authorize fraudulent transfers. Financial institutions encounter growing challenges as attackers use fake images and videos to bypass online security systems.
The latest International Safety Report indicates that threats from artificial intelligence-generated content will likely intensify. The report notes that AI-generated content is expected to become more persuasive due to improving capabilities, increased user dependence, and training on user feedback.
Detection Methods and Warning Signs
Levinzon explained that deepfake videos often contain detectable flaws. These include unnatural facial movements, inconsistent lighting or shadows, blurring, and distortion around the face. However, distinguishing authentic content from manipulated material grows more difficult as technology advances.
Services exist that analyze images and videos for signs of AI generation. These tools examine artifacts, unnatural textures, and errors in details. Levinzon noted that such tools provide probabilistic results rather than definitive verdicts.
Protection Strategies Against Deepfake Threats
Experts recommend limiting personal video content shared online, as criminals often use publicly available footage to create deepfakes. Multi-factor authentication for all accounts is essential, as compromised accounts may be used to post impersonating content.
Levinzon said VPN networks significantly strengthen security by encrypting internet traffic. This makes it more difficult for third parties, including internet service providers, to monitor online activity. Reduced data exposure lowers the risk of targeting by deepfake schemes and other fraud types.