AI to shape cyber threats: Cybersecurity predictions for 2025

Cyber threats evolved faster than before in 2024, primarily due to the rise and popularity of AI tools.

These tools assist cybercriminals in refining and making attack tactics more sophisticated and more difficult to recognise, while it helps cybersecurity professionals defend against attacks more effectively.

Cybersecurity

CEO of renowned cybersecurity platform KnowBe4 Stu Sjouwerman said the landscape is rapidly evolving, and the dynamic between defenders and attackers has never been more complex.

“As we enter 2025, we must embrace the potential of AI to enhance our defences and protect organisations globally.”

Sjouwerman said it is more important than ever to focus on the human element in organisations to lower the risk of becoming a victim of cybercrime.

With AI Advances KnowBe4 shared in its cybersecurity predictions for 2025

AI-enabled tools

Sophisticated AI-powered tools that detect and respond to threats more efficiently are being developed.

Capabilities like AI being able to analyse big amounts of data, identify anomalies, and enhance the accuracy of threat detection will be of massive assistance to cybersecurity teams going forward.

However, cybercriminals are also adopting AI to create more advanced attack methods.

As AI capabilities evolve on both sides, the standoff between defenders and attackers intensifies, making constant innovation and adaptation crucial.

ALSO READ: Data breaches cost SA organisations over R360m in 3 years

Ransomware attacks

Ransomware attacks will continue to be a massive threat due to the collaboration between ransomware gangs and initial access brokers.

In a measure to combat this, AI will become a popular tool to monitor networks and individual devices for anomalies like unusual encryption processes. This will greatly reduce the impact of attempted ransomware attacks.

Human factor

Organisations will continue to recognize the importance of frequent security awareness training and simulated phishing tests to manage the inherent human risk that exists within it, Sjouwerman predicts.

At the same time, cybercriminals will keep refining their social engineering techniques, making attacks more personalised and effective.

Going forward, the challenge will lie in maintaining employee vigilance without causing phishing fatigue.

To prevent this, it is important for organisations to focus on making training more adaptive and relevant to employees in order to create better protections and engagement for a positive security culture.

Deepfake detection

2025 will see deepfake AI detection technologies improve, become more accessible, and more effectively address the growing concern of identifying deepfakes.

On the other hand, cybercriminals are also expected to leverage disinformation and deepfakes in their attacks, using them to accelerate extortion, hide other attacks, or damage organisational reputations.

Zero-trust and cyber-mindfulness

There will be wider adoption of a zero-trust mindset and cyber-mindfulness, representing a proactive approach to cybersecurity. Organisations embracing these principles encourage a vigilant attitude among their employees, treating every user and device as a potential threat.

Exploitation and human trafficking in Africa

The region has experienced an increase of cybercriminals targeting its youth with false job opportunities which many times results in human trafficking.

The worry is that the overlap of cybercrime and human trafficking will continue to escalate in 2025.

Targeted attacks like romance and job scams will increase as cybercriminals become more adept at exploiting economic hardships and limited digital literacy.

ALSO READ: Google launches new AI-powered features for Android smartphones