"Generative Artificial Intelligence to Enhance Cyber Attacks"

With the development of technologies such as artificial intelligence (AI), technology is becoming increasingly intelligent, and so are cyber attacks. Google's latest cybersecurity forecast shows that the rise of AI brings new threats.

This Wednesday, Google released the "Google Cloud Cybersecurity Forecast 2024", a report compiled in collaboration with numerous Google Cloud security teams, aiming to delve into the network landscape of the coming year.

The report found that generative AI and large language models (LLMs) will be used for various cyber attacks, such as phishing, SMS, and other social engineering operations, with the goal of making voice and video content and materials appear more legitimate.

For example, when using generative AI, the flaws of phishing attacks, such as spelling errors, grammar mistakes, and lack of cultural background, will be harder to detect because it excels at mimicking natural language.

In other cases, attackers can provide LLM-generated legitimate content and produce modified versions that fit the attacker's target but maintain the same style as the original input.

The report also predicts that LLMs and other generative AI tools will receive increasing development and be offered as paid services to help attackers deploy attacks more efficiently with less effort.

However, AI or LLMs themselves are not doing anything wrong, as attackers are merely using generative AI to create content, such as drafting invoice reminders, which is not inherently malicious. However, attackers can exploit it for their malicious purposes.

Another potential AI threat is information manipulation. Attackers can simply use AI prompts to create fake news, fake phone calls, and deepfake photos and videos using generated AI models.

These operations may undermine public trust in news and online information, causing people to become more skeptical or no longer trust the news they receive.

The report states, "This may make it increasingly difficult for businesses and governments to communicate with the public in the near future."

Despite attackers using AI to enhance their attacks, network defenders can also leverage this technology to counter more advanced defenses.

"AI has already provided significant advantages to our network defenders, enabling them to enhance their capabilities, reduce workload, and better withstand threats," said Phil Venables, CEO of Google Cloud AI. "We expect that with defenders having this technology and guiding its development considering specific use cases, the capabilities and advantages brought by AI will skyrocket in 2024."

Defenders can use generative AI to generate actionable detections and take action at a faster pace.