eSentire is a leading provider of managed detection and response services, serving over 2,000 organizations globally in various industries. Their services focus on protecting users, data, and applications from cyber threats and enhancing overall security posture. To improve customer experiences, eSentire developed AI Investigator, a natural language query tool using generative AI capabilities on AWS.
AI Investigator enables customers to access security platform data through natural language queries, consolidating multiple data sources for seamless investigation. This tool has streamlined processes for over 100 analysts, enhancing investigation outcomes and reducing time. eSentire prioritizes privacy and security, hence developing its own large language model (LLM) rather than relying on third-party providers.
The AI solution was built using Amazon SageMaker, leveraging Amazon S3 for data storage and various SageMaker features for training and deployment. By customizing GPU instances based on workload complexity, eSentire optimized performance and cost-effectiveness. Moreover, a monitoring framework ensures visibility into AI interactions and maintains trust in the system.
Amazon Bedrock offers further flexibility and cost-efficiency for training custom models, enhancing eSentire’s AI capabilities. The LLM gateway project monitors interactions, ensuring security and compliance with user activity logging. AI Investigator has already provided significant time savings for customers and analysts by simplifying data exploration and response processes.
In conclusion, eSentire’s AI Investigator is a powerful tool that delivers quick and secure self-service interactions for customers’ complex data queries. This innovative solution demonstrates the company’s commitment to advancing cybersecurity through AI technologies.
Article Source
https://aws.amazon.com/blogs/machine-learning/esentire-delivers-private-and-secure-generative-ai-interactions-to-customers-with-amazon-sagemaker/