"How must databases adapt to generative AI, and how should databases be integrated with large language models (LLMs)? These are questions that Sailesh Krishnamurthy has grappled with for several years now. As VP of engineering for databases at Google Cloud, Krishnamurthy leads the database team for Google Cloud and all Google services including Google Search and YouTube. He also leads a program to leverage generative AI and Google's Gemini models in database management."
"The data is heavily permissioned and has to be secure. We worry about exfiltration and access. The data is at the heart of your line of business application, but it is also changing all the time, and if you keep extracting the data into some other corpus it gets stale. You can view it as two approaches: replication or federation."
LLMs provide broad world knowledge but must be combined with internal operational data for enterprise value. Extracting database records into external corpora creates staleness and exfiltration risks. Enterprise data requires strict permissions, security, and timely access. Integration options include replicating data into indexable stores or federating queries to live data. Generating accurate SQL from natural language is challenging and needs schema-aware, permission-aware tooling. AI-native database features—secure connectors, live federation, permission enforcement, and close LLM integration—are necessary to safely bridge generative AI and operational databases.
Read at InfoWorld
Unable to calculate read time
Collection
[
|
...
]