Right this moment, we’re saying Sec-Gemini v1, a brand new experimental AI mannequin centered on advancing cybersecurity AI frontiers.
As outlined a yr in the past, defenders face the daunting activity of securing towards all cyber threats, whereas attackers have to efficiently discover and exploit solely a single vulnerability. This elementary asymmetry has made securing techniques extraordinarily tough, time consuming and error susceptible. AI-powered cybersecurity workflows have the potential to assist shift the stability again to the defenders by pressure multiplying cybersecurity professionals like by no means earlier than.
Successfully powering SecOps workflows requires state-of-the-art reasoning capabilities and intensive present cybersecurity information. Sec-Gemini v1 achieves this by combining Gemini’s superior capabilities with close to real-time cybersecurity information and tooling. This mix permits it to realize superior efficiency on key cybersecurity workflows, together with incident root trigger evaluation, risk evaluation, and vulnerability affect understanding.
We firmly imagine that efficiently pushing AI cybersecurity frontiers to decisively tilt the stability in favor of the defenders requires a robust collaboration throughout the cybersecurity neighborhood. That is why we’re making Sec-Gemini v1 freely obtainable to pick out organizations, establishments, professionals, and NGOs for analysis functions.
Sec-Gemini v1 outperforms different fashions on key cybersecurity benchmarks on account of its superior integration of Google Menace Intelligence (GTI), OSV, and different key knowledge sources. Sec-Gemini v1 outperforms different fashions on CTI-MCQ, a number one risk intelligence benchmark, by at the least 11% (See Determine 1). It additionally outperforms different fashions by at the least 10.5% on the CTI-Root Trigger Mapping benchmark (See Determine 2):
Determine 1: Sec-Gemini v1 outperforms different fashions on the CTI-MCQ Cybersecurity Menace Intelligence benchmark.
Determine 2: Sec-Gemini v1 has outperformed different fashions in a Cybersecurity Menace Intelligence-Root Trigger Mapping (CTI-RCM) benchmark that evaluates an LLM’s potential to grasp the nuances of vulnerability descriptions, establish vulnerabilities underlying root causes, and precisely classify them in line with the CWE taxonomy.
Under is an instance of the comprehensiveness of Sec-Gemini v1’s solutions in response to key cybersecurity questions. First, Sec-Gemini v1 is ready to decide that Salt Storm is a risk actor (not all fashions do) and gives a complete description of that risk actor, due to its deep integration with Mandiant Menace intelligence knowledge.
Subsequent, in response to a query concerning the vulnerabilities within the Salt Storm description, Sec-Gemini v1 outputs not solely vulnerability particulars (due to its integration with OSV knowledge, the open-source vulnerabilities database operated by Google), but in addition contextualizes the vulnerabilities with respect to risk actors (utilizing Mandiant knowledge). With Sec-Gemini v1, analysts can perceive the danger and risk profile related to particular vulnerabilities sooner.
In case you are fascinated with collaborating with us on advancing the AI cybersecurity frontier, please request early entry to Sec-Gemini v1 through this manner.