As businesses race to implement artificial intelligence, many are discovering that the biggest challenge lies not in building the models but in ensuring that the underlying data is reliable and compliant. Without effective AI governance, companies risk making erroneous decisions and violating privacy regulations.
To address this issue, enterprise data intelligence provider Collibra has launched a comprehensive suite of new AI tools focused on governance, automation, and data democratization—essential elements for establishing trust in AI.
AI Governance: A New Era of Oversight
Among the new offerings is the AI Governance application, designed to serve as a centralized hub for managing diverse AI initiatives across organizations. This tool bridges the gap between data science teams, which prototype models, and risk and compliance stakeholders, who must provide approvals before these models can go into production.
"Our customers are excited about AI, but they are realizing that success involves much more than just the algorithms," said Collibra CEO Felix Van de Maele. "Data is central to AI, making strong data governance essential."
Van de Maele elaborated on the phenomenon of "shadow AI," where organizations experiment and prototype without adequate oversight. "A major challenge is ensuring that data scientists and engineers can access the right, compliant data."
Implementing AI governance necessitates a clear change management strategy and collaboration among multiple stakeholders, including legal and compliance teams. Van de Maele emphasized the growing importance of adhering to existing and emerging regulations.
The AI Governance tool is built on Collibra's ten-year investment in data governance, showcasing that the principles of data governance align closely with AI governance. It facilitates defining policies, roles, and responsibilities, creating a standardized workflow for registering, approving, documenting, and monitoring AI use cases, thus enhancing transparency and accountability in AI applications.
Collibra AI: Automating Data Management Tasks
Collibra AI, another significant feature of this release, utilizes large language models (LLMs) to automate tedious data management processes that typically require manual effort. One key function is generating descriptions and definitions for data assets, which is critical for effective data cataloging and governance.
Van de Maele described it as a "copilot approach," where the system generates draft descriptions for user review. "Having clear descriptions is vital for building trust and understanding around data."
Moreover, Collibra AI can create data quality rules based on natural language requests. For instance, it can translate specifications like "country codes must adhere to ISO standards" into enforceable technical rules.
Additionally, the system can parse SQL queries and business intelligence reports to create lineage graphs. This visibility allows organizations to track data flow, identify sources of issues, and assess the impact of data changes more effectively.
"A common issue arises when users want to understand how a number on a dashboard was calculated," noted Van de Maele. "Traceable lineage makes it easier for organizations to trust their data."
Democratizing Data Access with Collibra Data Notebook
To further improve data accessibility, Collibra has introduced the Data Notebook, enabling business users to easily search for and access trusted data assets. This tool simplifies the process of exploring datasets while ensuring that all governance workflows operate seamlessly in the background to maintain compliance and security.
"We’re providing users with one-click access to datasets while integrating governance workflows," said Van de Maele, likening the experience to shopping on Amazon, where finding and accessing products is quick and efficient.
Future Prospects and Challenges
Collibra aims to leverage its extensive experience in data governance to excel in the expanding field of AI governance. "We've been focused on data governance for 15 years, long before data catalogs gained popularity," remarked Van de Maele. "This positions us well to address the needs of organizations in implementing AI governance."
Nonetheless, Collibra faces competition from larger tech giants like Microsoft, IBM, AWS, and Google, all of which are also developing AI governance tools. With over $590 million raised from investors like Sequoia Capital and ICONIQ Capital, Collibra must continue to innovate and expand its capabilities to remain competitive.
Despite this challenge, Van de Maele is optimistic, citing that many of these tech companies are also Collibra's customers and partners. He believes that the company's specialized focus on data governance distinguishes it from the competition.
"Our mission is to transform how organizations leverage data, ultimately aiming to change the world with data," Van de Maele concluded. "If we can become an integral part of the enterprise AI landscape, ensuring responsible data stewardship, the potential opportunities are immense."
Ranking
