Data is the key to building modern AI workflows

Eric Newcomer, Principal Analyst at Intellyx, emphasizes the critical role of data in constructing effective AI workflows, particularly in the realm of security incident response. He highlights how generative AI (gen AI) can automate and optimize operational processes by using both public and private data to train large language models (LLMs). Public data aids in identifying workflow steps and preparing organizations, while private data enhances execution efficiency, reducing employee cognitive load and expediting business benefits. Gen AI can streamline incident response by automatically triaging events and suggesting actions based on quality private data, thus improving resolution time, especially for customer-impacting issues. The Tines Workbench is showcased as an example of an AI-powered tool that automates security workflows by integrating with Elastic Security to process alerts and collaborate with various security tools. Newcomer notes the importance of high-quality data in AI's effectiveness and suggests a targeted approach to workflow design to mitigate AI inaccuracies, ultimately reducing cognitive demands on human operators and maintaining customer trust.