Google Cloud has unveiled innovations to its AI-powered lakehouse architecture, enhancing performance and interoperability through features like BigLake Iceberg native storage, unified analytical and operational engines, and high-performance enhancements for BigQuery and Apache Spark. These advancements aim to empower organizations with better data management, analytics, and AI capabilities across their data ecosystems.

Google Cloud has introduced AI.GENERATE_TABLE, a new feature in BigQuery that enables the automated conversion of unstructured data into structured tables using advanced AI models like Gemini 2.5 Pro/Flash. This function streamlines data analysis by extracting key information from various formats, including images and medical transcriptions, and organizing it into a consistent schema for easier integration with existing workflows.

Google Cloud has launched a serverless version of Apache Spark integrated within BigQuery, aimed at simplifying data processing and analytics. This new offering eliminates the need for cluster management, reduces costs, and enhances performance while providing a unified development experience in BigQuery Studio, allowing users to seamlessly work with both Spark and BigQuery.