Artificial intelligence is transforming the way businesses operate, but the same technology is also creating new opportunities for cybercriminals. As AI becomes faster, more capable and increasingly autonomous, attackers can use it to automate activities that previously required significant technical expertise.
HKCERT has highlighted six emerging AI assisted attack methods that demonstrate how quickly the cybersecurity landscape is changing. The findings underline an important reality for businesses. AI is not simply another technology risk. It is becoming part of the attack process itself.
Agentic AI Is Changing Cyber Attacks
One of the most significant concerns identified by HKCERT is the rise of agentic AI. Unlike conventional AI tools that primarily generate information or recommendations, agentic systems can perform tasks and interact with computer environments.
Consequently, attackers may be able to use multiple AI agents to automate different stages of an intrusion. Reconnaissance, system analysis and other attack activities can potentially be coordinated with considerably less human involvement.
This development represents an important area within current AI trends and insights because greater autonomy could increase both the speed and scale of cyber threats.
AI Can Make CAPTCHA Protection Less Effective
CAPTCHAs have traditionally helped websites distinguish humans from automated programs. However, advances in AI based image analysis are making some traditional CAPTCHA systems easier for automated tools to solve.
As a result, organisations that continue relying on older verification methods may face additional risks from automated attacks. HKCERT recommends considering stronger interactive or behaviour based verification approaches.
This demonstrates how Machine learning advancements can create both defensive opportunities and unexpected vulnerabilities.
AI Assisted Website Attacks Can Scale Faster
Website reconnaissance and automated attacks can also become more efficient when AI is incorporated into the process. Attackers can potentially use AI systems to identify websites, locate login interfaces and coordinate attempts against multiple targets.
The ability to operate across numerous websites simultaneously creates a significant challenge for security teams. Therefore, organisations need strong authentication, multi factor authentication, regular vulnerability assessments and effective monitoring.
These changes are closely connected to broader Generative AI developments, where increasingly capable models can assist with complex technical tasks.
AI Is Reshaping DDoS Attacks
Distributed denial of service attacks have traditionally relied heavily on overwhelming systems with large volumes of traffic. However, AI can potentially make attacks more adaptive.
According to HKCERT, agentic AI could monitor defensive responses and adjust attack behaviour accordingly. This could make conventional approaches less effective because attackers may dynamically respond to rate limiting and other security controls.
At the same time, AI can strengthen defence by analysing network traffic and adjusting protective measures. Therefore, Automation and future tech could become an important part of both offensive and defensive cybersecurity strategies.
AI Driven Ransomware Raises New Concerns
Ransomware remains one of the most disruptive cyber threats, and AI could increase its adaptability. HKCERT discussed research into AI based ransomware capable of generating attack code dynamically rather than relying entirely on traditional prewritten malware.
This concept raises concerns because malicious software could potentially adapt itself to different environments. Consequently, security teams may need to combine endpoint protection with stronger access controls, network monitoring, backup strategies and employee awareness.
The development also highlights why responsible AI security must remain central to current AI industry updates.
AI Generated Phishing Websites Are Becoming More Convincing
Phishing remains one of the most common methods used to trick people into revealing sensitive information. AI is making the creation of convincing websites faster and easier.
Attackers can use AI assisted development tools to reproduce the appearance and functionality of legitimate websites. In some cases, websites can also be published quickly, reducing the time between creation and deployment.
Therefore, users can no longer rely only on how professional a website looks. Checking the URL, verifying unexpected requests and avoiding suspicious links remain essential security practices.
Why AI Security Requires a New Approach
The six threats identified by HKCERT demonstrate that AI is influencing cybersecurity at multiple levels. Attackers can use it to increase automation, improve scalability and adapt their activities, while defenders can use the same technology to detect suspicious behaviour and respond faster.
However, technology alone cannot solve every security challenge. Organisations also need clear AI governance, employee awareness, strong authentication and continuous security monitoring.
HKCERT’s broader cybersecurity outlook for 2026 reinforces this concern. The organisation reported 15,877 cybersecurity incidents in Hong Kong during 2025, representing a 27 percent annual increase, while identifying AI driven attacks and agentic AI risks among the major cybersecurity concerns for 2026.
What Businesses Can Learn From These Threats
The most important lesson is that organisations should treat AI security as an ongoing process rather than a one time technology project. Security teams should understand how AI tools are being used internally while also monitoring how attackers are adopting emerging capabilities.
Furthermore, businesses should strengthen authentication, update vulnerable systems, monitor unusual activity and educate employees about increasingly convincing phishing attempts. AI can also become part of the defensive strategy by helping security teams analyse large volumes of data and identify suspicious behaviour.
For organisations following Future of AI research, the security implications deserve as much attention as the productivity benefits. As AI becomes more autonomous, cybersecurity strategies will need to become equally adaptive.
Building a Stronger AI Security Strategy
The rise of AI assisted cyberattacks does not mean businesses should avoid AI. Instead, organisations should adopt it with appropriate safeguards and a clear understanding of its risks.
The six threats highlighted by HKCERT show that the next phase of cybersecurity will involve a continuous contest between increasingly automated attacks and increasingly intelligent defence systems. Businesses that combine technology, skilled people and strong security practices will be better positioned to respond as this landscape evolves.
Stay informed about the latest AI developments, cybersecurity risks and emerging technology trends with AITechInfoPro.
Reach out to AI Tech Info Pro for practical insights into responsible AI adoption, emerging cyber threats and the future of intelligent technology.
