Google’s cloud division has followed in Microsoft’s footsteps with the launch of Security AI Workbench, which leverages generative AI models to gain better visibility into the threat landscape.
Powering the cybersecurity suite is Sec-PaLM, a specialized large-scale language model (LLM) “fine-tuned for security use cases.”
It leverages the latest advances in AI to enhance point-in-time incident analysis, threat detection, and analytics to deliver trusted, relevant, and actionable intelligence to help It is intended to combat and prevent new infections.
To that end, Security AI Workbench spans a wide range of new AI-powered tools, including VirusTotal Code Insight and Mandiant Breach Analytics for Chronicle, to analyze potentially malicious scripts and alert customers to active breaches in their environment. To do.
Similar to Microsoft’s GPT-4-based Security Copilot, users will be able to “search, analyze, and investigate security data in a conversational manner” with the goal of reducing mean response times and quickly determining the full range of events. “can.
Meanwhile, VirusTotal’s Code Insight feature is designed to generate natural-language summaries of code snippets to detect and mitigate potential threats. It can also be used to flag false negatives and clear false positives.
Another important product is Security Command Center AI. It leverages Sec-PaLM to provide operators with “near-instantaneous analysis of findings and possible attack paths,” affected assets, and recommended mitigations.
Google also uses machine learning models to detect and respond to API abuse and business logic attacks. Attackers weaponize legitimate functions to achieve malicious goals without triggering security alerts.
Zero Trust + Deception: Learn How to Outsmart Attackers!
See how Deception can detect advanced threats, stop lateral movement, and strengthen your Zero Trust strategy. Join us for an insightful webinar!
Save my seat!
“Security AI Workbench is built on Google Cloud’s Vertex AI infrastructure, enabling customers to manage their data with enterprise-level features such as data segregation, data protection, sovereignty, and compliance support,” said Google. Cloud’s Sunil Potti said.
The development is part of a new unit called Google DeepMind, in which Google brings together DeepMind’s AI research group and Google Research’s Brain team to “build safer, more responsible, and more capable systems.” It takes place a few days after announcing its creation.
News of Google’s Security AI Workbench also follows a plan to integrate AI into GitLab’s platform to help developers prevent access tokens from being leaked and avoid false positives during security testing. increase.
