Read More>> 
According to the recent ‘State of the UAE Cybersecurity Report 2025’ released by a G42 company from the UAE, more than 223,800 digital assets hosted in the UAE are potentially exposed to cyberattacks. Adding to the woes, half of the critical vulnerabilities remain unaddressed for over a few years.
This report also examines the distinct cybersecurity challenges confronting the UAE, highlighting the rise of AI-driven threats, cybercriminals’ evolving strategies, and the increasing use of AI by state-sponsored actors in advanced persistent threats (APTs).
Some notable incidents of AI-driven cyber-attacks globally include –
- Deepfake CEO Fraud (2019) – Criminals used AI-generated audio to impersonate a CEO, tricking an employee into transferring $243,000
- BlackMamba (2023) – A proof-of-concept malware that used AI to generate malicious code in real-time, making it nearly undetectable
- WormGPT & FraudGPT (2023) – AI tools found on dark web forums designed to generate phishing emails, malware, and fraud scripts
‘Why’ and ‘How’ of AI-driven attacks
As AI advances, cybercriminals also leverage their power to conduct sophisticated cyberattacks. AI-driven attacks pose new challenges for cybersecurity professionals, making traditional security measures increasingly inadequate. Attackers are exploiting AI to increase the effectiveness, automation, and scale of cyber threats.
In this context, it has been observed that AI-driven attacks happen because cybercriminals leverage artificial intelligence to automate, enhance, and scale their attacks. AI can analyze vast amounts of data, identify vulnerabilities, bypass security measures, and generate convincing phishing messages or deepfakes. Attackers use AI to optimize malware, conduct advanced social engineering, and evade detection, making cyber threats more sophisticated and difficult to counter. They employ AI for tasks like:
- Phishing (creating highly personalized messages)
- Malware development (adapting to security defenses)
- Deepfake generation (manipulating audio and video)
- Vulnerability exploitation (quickly identifying weak points in systems)
Mitigation of AI-based cyber risks
Numerous brainstorms have been conducted on the mitigation of AI-based cyber threats. While experiments are still being done, some evident strategies can address threats arising from AI-based technologies. E.g.
- Launch cybersecurity awareness and education initiatives to train employees on AI-based threats
- Run regular cybersecurity audits and compliance checks to detect anomalies
- Create an asset inventory to identify network anomalies
- Set up a 24/7 Security Operations Centre for monitoring and analysis of every activity
- Implement Endpoint Detection and Response (EDR)
- Establish a robust cyber threat intelligence function
- Develop/ follow AI governance frameworks for the safe and ethical use of AI
The Bottom-Line
AI-driven cyberattacks are no longer a futuristic concern — they are here today, evolving rapidly and challenging conventional security defenses. Organizations must embrace AI-driven security strategies, adopt proactive defense measures, and foster a culture of cybersecurity awareness to mitigate risks in this era of cyber warfare.
