Sweet now harnesses generative AI to nail business-critical cloud risks and reduce cloud Mean-Time-To-Respond (MTTR) and Mean-Time-To-Contain (MTTC)
Sweet Security announced it has added generative artificial intelligence (GenAI)-powered capabilities to its cloud runtime security suite. Sweet users now have an upgraded ability for runtime risk management and runtime response as they can classify workloads by business impact and get granular incident response playbooks on the fly. By integrating GenAI, Sweet enables security teams to act on its runtime insights faster and reach Mean-Time-To-Contain (MTTC) quicker than ever before. Sweet will be demonstrating its new AI-powered capabilities at Black Hat, booth #3113.
Mandiant’s 2024 m-trends report underscores the importance of effective response to cyber attacks, stating those with an existing incident response plan and broad environmental monitoring are better prepared. Sweet’s AI-generated response plans ensure that SOC teams are well-prepared with detailed instructions on how to quickly and safely intervene when an incident occurs, in addition to insight on attacker tactics, damage assessments, and the full attack path taken by the threat actor.
Runtime Insights + GenAI = Lean, Mean, Response Teams
Sweet’s holistic approach to cloud security leverages runtime insights to deliver comprehensive protection across all layers of the stack—from cloud infrastructure assets, applications, secrets, identities, APIs, and network interactions (Layer 7). This enables it to detect sophisticated attacks that evade siloed detection solutions. But complex risks require business focus and agility — resources in short supply when dealing with active threats.
Also Read: Radware Launches EPIC-AI to Deliver New AI-Powered Intelligence and GenAI Capabilities Across its Security Solutions and Services
Sweet’s AI-powered risk management and response framework help security teams nail business-critical risks and reduce the Mean Time to Respond (MTTR) for incidents with the following new features:
Workload classification — for better vulnerability prioritization: Using GenAI, security teams can now classify workloads by business criticality. Adding to Sweet’s existing vulnerabilities filtering feature, which showcases if a vulnerability is public facing/loaded/executed, etc., security teams can now understand if the workload/s it impacts is business critical. This enables security teams to prioritize the 1% of active risks still in the queue.
Dynamically generated response playbooks — to expedite attack response and recovery: Sweet’s AI-powered response playbooks offer step-by-step instructions cyber analysts can leverage to quickly get to the bottom of the incident, collect the right artifacts needed for decision-making, and take action to prevent escalation.
Sweet’s GenAI – how it works under the hood
Sweet’s Large Language Models (LLMs) rely solely on models deployed internally, so private data is not exposed to third-party services. The LLMs are specifically trained with technical data across cybersecurity, DevOps, and other relevant domains, taking in a very broad context as an input to provide outputs that are hyper relevant to the specific environment and situation, such as the concrete code snippets to run. Once the context has been properly aggregated, the next step is to break down the big use-case to a chain of smaller increments that the LLM can correctly handle without digressing into hallucinations.
“They say, attackers only need to be right once, but defenders need to be right 100% of the time, but well-applied GenAI can level the playing field,” said Eyal Fisher, co-founder and chief product officer, Sweet Security. “That was our guiding light when we chose vulnerability management and incident response as our initial use cases. We are proud to provide AI functionality that enables security teams to stay ahead of an adversary, because that’s how you shut down an attack.”
Source: PRNewswire
