Granica is an AI data platform built to make enterprise training data safer and more cost-efficient. It operationalizes a data-centric approach to AI through two main capabilities: large-scale data reduction that lowers cloud storage and processing costs, and real-time sensitive-data discovery, classification, and masking for both data lakes and LLM prompts.
The privacy side, branded around protecting PII and other sensitive information, lets AI teams using cloud object storage like Amazon S3 and Google Cloud Storage discover and mask sensitive data before it is used in training or sent to LLMs. This addresses a top enterprise concern: preventing sensitive data exposure as AI pipelines proliferate.
Granica emerged from stealth in June 2023 with $45M in Series A funding from New Enterprise Associates (NEA) and Bain Capital Ventures, with participation from Abstract Ventures, Uncorrelated Ventures, Original Capital, K9 Ventures, and angels including former Tesla CFO Deepak Ahuja and Okta co-founder Frederic Kerrest.
The company serves AI/ML and data platform teams at enterprises that need to control the cost and privacy risk of large training datasets. By reducing data volume and masking sensitive fields, it aims to improve AI ROI while supporting compliance.
Granica sits at the intersection of AI privacy and data engineering/infrastructure, addressing the data-layer challenges of building AI responsibly at scale.