How businesses are fighting sophisticated cyber threats with AI

Techday NZ

14-05-2025

In an era where artificial intelligence (AI) is redefining the corporate landscape, its darker side is also beginning to take shape - and it's hitting Australian and New Zealand organisations where it hurts most: cybersecurity.
Ransomware attacks, already a formidable threat to businesses, are being supercharged by AI. Threat actors are leveraging the technology not just to improve existing methods but to innovate entirely new tactics.
The result is a new class of cyberattacks—faster, more convincing, and increasingly difficult to detect.
An escalating threat
AI is accelerating the capabilities of ransomware operators. While the core mechanics of a ransomware attack remain largely unchanged (encrypting data and demanding payment) AI enhances each phase of the attack lifecycle.
According to cybersecurity experts, the most alarming consequence is the lowered barrier to entry for cybercriminals. Anyone with access to the right AI tools can now generate malware, craft convincing phishing emails, or exploit known vulnerabilities. They no longer need deep technical knowledge.
AI-generated phishing emails and deepfake voice scams are becoming alarmingly realistic. Attackers can automate these campaigns at scale, fooling even vigilant employees.
Traditional safeguards, such as spam filters and antivirus programs, are no match for the adaptive and evolving nature of these AI-enhanced threats.
The speed of attack is also increasing. AI algorithms can scan networks, identify vulnerabilities, and develop targeted exploits in a fraction of the time it would take a human operator. As a result, businesses have a shrinking window to detect and patch weaknesses before they are exploited.
A tipping point for cybersecurity
The emergence of AI-driven attacks is pushing many organisations to rethink their entire approach to cybersecurity. Signature-based detection systems, long a staple in enterprise environments, are proving inadequate against threats that constantly shift form.
The reality is that these attacks are becoming too intelligent for reactive systems. If an organisation is relying on known signatures or pre-defined rules, it will always be a step behind.
This urgency is driving a growing shift toward analytics-driven security strategies - particularly User and Entity Behaviour Analytics (UEBA). This AI-powered approach focuses not on identifying known threats, but on recognising unusual patterns in user and system behaviour.
Using AI to fight AI
UEBA is emerging as a vital tool in the defence arsenal. It leverages machine learning, behavioural analytics, and data science to detect anomalies that could indicate malicious activity, even when that activity doesn't match any known threat profile.
One of UEBA's key strengths lies in its ability to proactively hunt for threats. It scans for patterns that deviate from the norm, such as an employee logging in from an unusual location or accessing systems at odd hours.
By doing so, it offers early warnings of potential breaches and speeds up investigation times for security teams.
Automation is another powerful component of UEBA. The technology not only identifies threats but can also initiate pre-defined response actions automatically. Combined with orchestration tools, this reduces the burden on Security Operations Centres (SOCs) and allows for a faster, more coordinated response.
Accuracy is also improving. Unlike legacy systems that generate a high number of false positives, UEBA tools assign risk scores to anomalies, allowing teams to focus on high-probability threats. Over time, the system becomes more precise, adapting to an organisation's unique behavioural baseline.
A strategic imperative
The growing complexity of ransomware and phishing attacks is making cybersecurity not just a technical concern, but a strategic business priority. For sectors like finance, healthcare, and critical infrastructure - where data breaches can have regulatory, reputational, and financial consequences - the stakes are particularly high.
Many Australian and New Zealand firms are already investing heavily in AI-based defence technologies. But experts warn that implementation alone isn't enough. A comprehensive, security-first culture is essential and should include employee education, executive oversight, and alignment with broader risk management strategies.
As the arms race between cybercriminals and defenders escalates, organisations must recognise that artificial intelligence is not just a threat - it's also the best line of defence. By deploying AI to combat AI, businesses can take a proactive stance in an increasingly hostile digital environment.
The security landscape has changed. It's no longer a case of just defending against people behind keyboards. It's also about defending against intelligent systems. To win, organisations need smarter defences, powered by AI, informed by data, and driven by strategy.