As cyber threats become faster, more complex, and increasingly automated, artificial intelligence (AI) is stepping in as the most powerful ally in modern cybersecurity defence.
In 2025, the integration of AI and machine learning (ML) into cybersecurity operations is no longer experimental—it is essential.
From real-time threat detection to automating Security Operations Center (SOC) workflows, AI is not only enhancing the efficiency of cybersecurity teams but is also filling critical gaps in the cybersecurity workforce, especially in countries like Malaysia where demand for skilled professionals outpaces supply.
The Shift to AI-Powered Threat Detection
Traditional signature-based antivirus tools are proving inadequate in an era of polymorphic malware and advanced persistent threats (APTs). AI-powered systems, such as those offered by Stellar Cyber and SentinelOne, use behavioural analytics to identify anomalies rather than relying on known attack patterns.
Stellar Cyber, for instance, utilises Open XDR (Extended Detection and Response) that ingests data from across an organisation’s infrastructure, applies machine learning models to detect abnormal behaviour, and automates responses—all within seconds (Stellar Cyber, 2025).
Similarly, SentinelOne’s Singularity platform uses AI to analyse billions of telemetry data points to autonomously detect, investigate, and respond to threats across endpoints, containers, and the cloud (SentinelOne, 2025).
AI Reduces Time-to-Detection and Response
The average time to detect a breach in traditional systems was over 200 days as of 2022, according to IBM. In contrast, AI-driven platforms can reduce this to mere minutes. This speed is critical for limiting the damage of ransomware, insider threats, or zero-day exploits.
According to a 2024 report by Capgemini, 69% of organisations believe they will not be able to respond to cyberattacks without the support of AI. The report also highlights that companies using AI experienced a 12% improvement in threat detection accuracy and a 15% reduction in the cost of breach response.
Real-World Use Cases in Malaysia
Several Malaysian enterprises and government-linked companies (GLCs) are adopting AI-based solutions to secure critical infrastructure.
Telekom Malaysia, for example, has begun piloting AI-enhanced anomaly detection within its core network infrastructure, while local banks have started integrating AI-powered fraud detection engines that monitor millions of transactions in real time.
The Malaysian Communications and Multimedia Commission (MCMC) has encouraged organisations to adopt AI in cybersecurity, especially for protecting 5G infrastructure and smart city deployments, where the attack surface is vast and complex.
Automated SOCs: AI as a Force Multiplier
AI is transforming the traditional SOC into a more agile, responsive, and data-driven environment. AI automates tasks like log analysis, threat correlation, incident prioritisation, and even remediation actions such as isolating infected endpoints.
According to a Gartner forecast, by 2026, more than 60% of SOC operations will rely on AI and machine learning to function efficiently—particularly in hybrid or cloud-native infrastructures.
This automation is particularly valuable in Malaysia where SOC analyst burnout and skill shortages remain critical issues. Automated platforms enable smaller teams to manage larger attack surfaces, especially in SMEs with limited budgets.
The Ethics of Autonomous Cyber Defence
However, as with all AI applications, ethical considerations remain crucial. False positives, model bias, and over-reliance on automation can introduce new risks. Moreover, attackers are also leveraging AI to build more convincing phishing campaigns and evade detection.
Thus, industry best practices stress the need for human oversight in automated systems. Tools like SentinelOne and Stellar Cyber include mechanisms for analyst feedback to improve detection models continuously.
Malaysia’s National AI Roadmap 2021–2025 also emphasizes the importance of “human-in-the-loop” principles to ensure ethical, accountable AI deployment in critical sectors.
Conclusion: From Defence to Prediction
AI is moving cybersecurity from a defensive posture to a predictive one. Instead of merely reacting to incidents, organisations are now using AI to anticipate potential threats, assess vulnerabilities, and proactively strengthen defences.
For Malaysian businesses—especially those in finance, healthcare, logistics, and telecommunications—investing in AI-powered cybersecurity is no longer a luxury, but a necessity for survival in a threat landscape that evolves by the second.
Final Thoughts
Cybercriminals are evolving—and so should your cybersecurity. AI isn’t just a tool; it’s your frontline defence in a rapidly shifting digital battlefield.
Modern threats demand smarter, faster, and predictive solutions. With AI-powered platforms, you’re not just reacting to attacks—you’re preventing them before they start.
🛡️ Don’t wait for your employees to be the last line of defence.
👉 Start with Exabytes eSecure to explore how we can help you with cybersecurity-related issues.
References
- Capgemini. (2024). Reinventing Cybersecurity with Artificial Intelligence.
- Gartner. (2025). Cybersecurity Trends: Resilience Through Transformation.
- IBM. (2022). Cost of a Data Breach Report 2022.
- MCMC. (2025). Cybersecurity Guidelines for 5G and Smart Infrastructure.
- MyDigital. (2025). National Artificial Intelligence Roadmap 2021–2025.
- SentinelOne. (2025). Singularity XDR Platform Overview.
- Stellar Cyber. (2025). Open XDR Explained.
