Ataccama has launched its new AI Agent for data management, “a significant leap forward in autonomous AI”.
Ataccama ONE AI Agent, part of Ataccama’s data trust platform, is the first of its kind AI Agent aimed specifically at data management, capable of independently executing complex tasks with minimal guidance. The agent intelligently chooses the best approach for assigned tasks, self-corrects as necessary, manages complex data requirements independently and delivers outputs for review.
Speaking to TechBullion, Jay Limburn, Chief Product Officer for Ataccama explained what sets an AI Agent apart from assistants and copilots:
“AI agents are designed to operate autonomously to achieve specific tasks or goals and can take on more complex assignments. They often interact with digital systems, APIs or other agents to gather information, execute tasks and provide results. They are able to self-correct if something breaks along the execution journey, adjusting the steps and iterating on it before providing the result.”
Whereas AI assistants assist and augment human capabilities with low agency and autonomy, and AI copilots are primarily designed to provide decision support and operate like virtual assistants to streamline and speed up tasks, according to Limburn.
Designed for data engineers and business users, Ataccama’s AI Agent reduces manual touch points for data quality assurance, expedites data cleaning and reduces operational costs for data management tasks, providing clean, trustworthy data to enterprises. It aims to make data more accessible to business users, requiring no technical knowledge.
“Agentic AI is the next milestone in the evolution of AI, but for these intelligent systems to be capable of making decisions and taking actions autonomously, there needs to be trust in data,” said Limburn. “Agents within enterprises will enable AI systems to interact and collaborate, and data trust will become paramount for effective cooperation and problem-solving.”
Ataccama is delivering the data trust bridge needed between data sources and data consumers with our Ataccama ONE unified data trust platform which offers data quality, observability, lineage, governance and master data management in a single solution. Purpose-built AI tailored for specific tasks like data management will boost efficiency, support ethical interactions among agents, and facilitate human and AI collaboration to uphold industry standards and ensure trustworthy outcomes.”
Ataccama’s AI Agent has been trained on data management expertise and best practices, the Ataccama ONE unified data trust platform, working with the power of the ChatGPT 4 model to integrate external knowledge for it to adapt and responsd to a range of tasks.
It acts as a dedicated data companion to automate a wide range of data quality configurations including bulk data quality rule creation, intelligent data quality rule mapping and automated data quality evaluation, and answer queries for finding data assets. All AI-generated results are accompanied by a clear audit trail, allowing staff to identify where data has come from and spot potential issues.
The recent spell of companies launching AI Agents prompts questions over who governs these systems and the data they use.
Limburn said: “This often sits within the remit principally of the Chief Data Officer and secondarily with the Chief Technology Officers. Technology within the enterprise is fast evolving and in response, we are seeing a shift in responsibilities and leadership with the emergence of blended roles including Chief Data and AI Officer positions. We would expect these roles to be directly responsible for spearheading AI initiatives and the governance of agents as part of that in collaboration with other leadership roles.”
Ataccama said it’s looking to continue to iterate and build out its AI Agent features to scale its capabilities.
Ataccama has been a leading AI-driven data management provider since 2016, most recently recognised as an Emerging Specialist in the Gartner Innovation Guide to Generative AI October 2024.
