Google Uses Gemini Model in Cyber ​​Security

Google aims to make threat reports more understandable by directing artificial intelligence to the field of cyber security. The new cybersecurity product, Google Threat Intelligence, combines threat intelligence from Mandiant and VirusTotal with the Gemini AI model. This approach increases users’ security by providing faster and more accurate understanding of threats.

Features of Gemini 1.5 Pro

The new product uses the large tongue model Gemini 1.5 Pro. Google states that this model accelerates the reverse engineering of malware. For example, the fact that it detected the kill switch of the WannaCry virus by analyzing its code in just 34 seconds reveals the power of this model.

Another possible use of Gemini in the threat space is to summarize threat reports into natural language within Threat Intelligence. In this way, companies can understand the impact of possible attacks more clearly and avoid overreacting or underreacting. Organizations equipped with the right information become better prepared against cyber threats.

Threat Intelligence has an extensive information network to monitor potential threats before attacks. It allows users to see the bigger picture of the cybersecurity environment and prioritizes issues that need focus. Mandiant’s experts work to block attacks by monitoring malicious groups and collaborating with companies. The VirusTotal community also regularly shares threat indicators.

Google acquired Mandiant in 2022, which exposed the SolarWinds cyberattack against the US federal government in 2020. The company plans to evaluate the vulnerabilities of AI projects using Mandiant’s experts. Through Google’s Secure AI Framework, Mandiant will test the defenses of AI models and assist with red team efforts. AI models can summarize threats and reverse engineer malware attacks, but they can themselves be targeted by malicious actors.

Of course, Google is not the only company combining artificial intelligence with cybersecurity. Microsoft introduced Copilot for Security, powered by GPT-4 and Microsoft’s cybersecurity-focused AI model. This tool allows cybersecurity experts to ask questions about threats. It will be interesting to see how AI will revolutionize cybersecurity, although it is not yet clear which company will best use AI in this field.