Supercharge security with AI
Imagine a world where you have contextualized and actionable visibility into the latest threats, a world where security configurations are automatically created, and where even non-experts can detect, investigate, and respond to cyberthreats with confidence.
We're building this world with Duet AI in Google Cloud.
Duet AI in Security Operations is now generally available.
Help detect, contain, and stop threats from spreading
With our AI advances, you can now combine world-class threat intelligence with novel AI-based detections and analytics to identify and contain initial infections and help prevent them from spreading.
Duet AI in Mandiant Threat Intelligence offers AI-powered capabilities that help assess, summarize, and prioritize threat data across proprietary and public sources.
VirusTotal Code Insight helps analyze and explain the behavior of potentially malicious code without needing to reverse engineer scripts.
Help reduce toil and manual work
Duet AI can reduce much of the repetitive work that plagues cybersecurity practitioners. We are helping reduce the toil it takes to manage multiple environments, investigate incidents, and respond in a timely manner.
Duet AI in Security Operations enables you to search vast amounts of data in seconds with custom queries generated from natural language. Reduce time-consuming manual reviews by leveraging automatic summaries of case data and alerts, and improve response time with recommendations for the next steps to support incident remediation.
Simplify security for experts and non-experts alike
We anticipate that our generative AI will profoundly change how practitioners across skill levels “do” security. AI assistive features can help teams stay one step ahead of adversaries with near-instant analysis of security findings to help usher in a new era of effectiveness.
Duet AI in Security Command Center translates complex attack graphs to plain text explanations of exposure and simulates possible attack paths, highlighting impacted assets and recommending mitigations before assets can be exploited.
Google Cloud Security AI Workbench
Built on Vertex AI infrastructure and leveraging threat intelligence from Mandiant, VirusTotal, and Google, Security AI Workbench gives defenders more natural, creative, and effective ways to keep their organizations safe like never before.
- Powered by Sec-PaLM 2, a specialized security large language model (LLM)
- Fine-tuned for security use cases, incorporating threat intelligence
- Features extensions that allow customers and partners to build on top of the platform while still keeping control and isolation over their data
- Provides enterprise-grade data security and compliance support
Leading the charge in responsible AI
We’re committed to responsible AI. Customers can control their data with enterprise-grade capabilities such as data isolation, data protection, and compliance support.
"Chronicle has unlocked the value of cybersecurity data at a massive scale. We are excited about Duet AI’s potential to dramatically increase the effectiveness of our team. The ability to use natural language queries to perform complex analysis will help new analysts on-board faster and let seasoned analysts rapidly pursue advanced threats across our complex environment."
Guy Delp, Vice President, Global Information Security, Pfizer
Frequently asked questions
When will these offerings be available?
Duet AI in Security Operations is now generally available. VirusTotal Code Insight is also available now, putting AI to work for our customers.
We will be rolling out other offerings to trusted testers in coming months. Duet AI in Mandiant Threat Intelligence and Duet AI in Security Command Center are in Preview.
Which LLM is being used for the Security AI Workbench?
Google Cloud Security AI Workbench uses a specialized security LLM, Sec-PaLM 2. The security model is trained on PaLM 2 and fine-tuned for security use cases, incorporating our visibility into the threat landscape and Mandiant’s frontline intelligence on vulnerabilities, malware, threat indicators, and behavioral threat actor profiles.
What are your data practices when it comes to privacy? Will you be using customer data to train these models?
Our platform lets customers make their private data available to the platform at inference time; in doing this, we honor our data privacy commitments to customers. This feature helps ensure you control your data. Because Security AI Workbench is built on Vertex AI infrastructure, customers can get enterprise-grade capabilities such as strong data isolation, data protection, and compliance support.