The combination of AI developments and cybersecurity dangers has received attention as Australia commemorates its annual 'AI Month.' Reuben Koh, Akamai Technologies' Director of Security Strategy for Asia Pacific and Japan, has shared insights into how businesses may protect themselves from AI-related attacks.
Reuben Koh states, "Artificial intelligence (AI) is reshaping how we live, work, and play, with profound and complex implications for cybersecurity." It is more vital than ever for businesses to upgrade their security strategy, including defending AI systems, fighting AI-driven attacks, and maintaining data security for AI-Ops."
Areas of Focus
Koh identified two significant focus areas at the meeting: cybersecurity and artificial intelligence.
The first aspect is concerned with securing AI systems. The expanding use of AI in several industries creates new security challenges, forcing the implementation of stronger security measures. He identified three significant security issues related to AI systems.
- The first is AI data poisoning, in which attackers intentionally alter the AI's training data, potentially leading to unfair or harmful results.
- The second risk is quick injection. This occurs when attackers use created data to manipulate AI models, such as chatbots, into doing undesired activities, perhaps evading safety controls.
- The final risk pertains to data privacy. AI models require a large amount of data, particularly sensitive information, which creates worries about data security, misconfiguration, and improper access throughout the AI process.
The second area of concern is protecting against AI-powered threats. As hackers use AI to improve their methods, companies must plan to secure their systems against increasingly sophisticated AI-driven threats.
According to Koh, AI-enhanced malware is becoming increasingly clever and mysterious, making it more difficult to identify and halt.
AI Threat
Furthermore, AI-driven social engineering poses a threat, as AI technologies generate highly convincing phishing scams, including deepfake videos and sounds, raising the risk of such attacks.
AI automation of major segments of the attack chain improves the efficiency of attacks by giving cybercriminals less time to exploit dangers, reducing the opportunity for incident reaction by defenders.
According to the report, cybersecurity breaches at Australian organizations, such as ransomware operations and supply chain attacks, are becoming increasingly influential. The most obvious have been customer data hacks at financial institutions and attacks on critical telecommunications and healthcare infrastructure. As the Australian government has implemented new cybersecurity laws and rules and modernized its defenses, businesses have increased their focus on data privacy and security.
Koh Statement
Koh stated as:
"The use of AI can and will inevitably lead to increased risks to security. At the same time, the number of AI-powered attacks is growing as the technology matures and gets more widely used. As AI becomes more common, companies must prioritize learning how to defend against AI threats and secure AI systems."