The intelligent era is upon us, bringing remarkable innovation and an increasingly complex cybersecurity landscape. As AI becomes woven into our digital fabric, it serves as both a catalyst for progress and a potential weapon in the wrong hands. Navigating this new frontier requires fundamentally rethinking security—moving beyond traditional defences toward intelligent, adaptive systems. At its core, this challenge is about trust: ensuring our increasingly digital world becomes more secure, not less.
AI is rapidly transforming cyber threats. Adversaries now deploy sophisticated algorithms to automate attacks, craft convincing phishing campaigns, and develop adaptive malware. The result is a 50% year-over-year increase in AI-driven network attacks in 2024. Yet this same technology offers powerful defensive capabilities. AI’s ability to analyse vast datasets, spot subtle anomalies, and predict threats in real-time has become essential for security professionals trying to stay ahead of increasingly sophisticated attackers.
Defending against AI-augmented threats requires a fundamental evolution in cybersecurity thinking. The answer lies in AI-Native security, where artificial intelligence forms the foundation of security architecture rather than serving as an add-on tool. This approach demands systems designed from inception to be intelligent, automated, and self-improving.
AI-native security: A necessary evolution
As Dr. Zhu Shenggao, Vice President of AI at Huawei Cloud Middle East & Central Asia, explained at a recent media briefing at GISEC GLOBAL 2025, “As AI models grow, Huawei Cloud has evolved from cloud-native to AI-native security. Huawei Cloud now puts AI at the heart of its strategy. With our AI-native approach, we help organisations stay ahead of evolving threats using smart, self-improving security systems.”
This vision is represented by innovations like Huawei’s Pangu AI security models, which leverage deep learning and vast security datasets. These systems demonstrate AI-Native security’s potential by automating 99% of threat responses while dramatically reducing incident detection and attack source tracing times.
Similarly, integrated defence strategies are essential in today’s complex threat environment. Frameworks like Huawei’s ‘one centre, seven defences’ model, which provides comprehensive protection across networks, applications, and data, become increasingly critical as attack surfaces expand. AI’s transformative power shines in its ability to unify these protection layers, providing intelligent oversight and enabling proactive defence.
Dr. Zhu noted, “In data protection, these models will enhance the accuracy and efficiency of data classification and sensitive data identification. They will also help detect abnormal data access behaviours and data leakage risks.”
Examples of such integrated defence solutions include the Huawei Xinghe Intelligent Unified SASE (Secure Access Service Edge) solution, which combines cloud, network, edge, and endpoint security under a unified AI-driven security brain. This enables consistent policy enforcement and threat intelligence correlation across distributed environments. At the endpoint level, products such as HiSec Endpoint leverage AI for advanced threat detection and behaviour analysis and can even trigger proactive measures like file backup when ransomware-like encryption activity is detected.
Securing AI itself: The next critical frontier
As organisations increasingly rely on AI for security, a new imperative emerges: ensuring the security and integrity of AI models themselves. Dr. Zhu highlighted that “large AI models face targeted attacks and ethical risks, making security, compliance, privacy, reliability, and control critical priorities for businesses.” If AI systems are to be the guardians of our digital assets, they must be robust, reliable, compliant, and ethically sound.
This involves protecting AI models from adversarial attacks, ensuring the quality and privacy of their training data, and guaranteeing trustworthy outputs. Innovative concepts like “using AI models for AI model security,” as proposed by Huawei Cloud, point toward a future where AI not only defends our systems but also safeguards its own operational integrity across environment protection, data security, model reliability, and content safety.
Trust and collaboration: The bedrock of digital innovation
Ultimately, the journey toward a secure intelligent era depends on establishing trust. “Cybersecurity and privacy protection are the cornerstones of development in the digital and intelligent world,” Dr. Zhu noted. This trust is built through advanced technology, transparency, adherence to global standards, and robust governance frameworks.
Furthermore, compliance with comprehensive standards like Huawei Cloud’s Cloud Service Cybersecurity & Compliance Standard (3CS) – which draws from more than 16 global security standards and 30 years of expertise – is essential for navigating complex international regulations and ensuring data sovereignty across diverse deployments.
The rise of artificial intelligence presents both extraordinary opportunities and significant cybersecurity challenges. Creating a secure path forward requires a proactive, intelligent, and holistic security posture. By embracing AI-Native security principles, driving innovation in areas like AI model security, and committing to collaborative frameworks built on trust and transparency, we can harness AI’s transformative power while effectively managing its risks. This concerted effort, supported by comprehensive solutions and expert insights, will pave the way for a truly connected and resilient intelligent world.
Discussion about this post