Zürich, Switzerland – September 18, 2025 – Squirro, a global leader in enterprise-grade Generative AI and knowledge graph solutions, today announced the general availability of its latest platform update. This introduces powerful new capabilities, including the Squirro Classifier, which streamlines data classification against an organization’s enterprise taxonomy. Further additions include advanced personally identifiable information (PII) detection, masking for enhanced data privacy, and support for the model context protocol (MCP). These features will be invaluable in helping organizations across all industries unlock more accurate, secure, and context-aware insights from their unstructured data.
“We're at an interesting inflection point in the GenAI landscape. With the pace of incremental improvements in foundational models slowing down, the real differentiator is shifting towards how the technology can deliver value in the enterprise," said Dorian Selz, CEO of Squirro. "Our deep expertise lies in making the benefits of GenAI accessible to enterprise users who value security, data privacy, accuracy, and scalability.
Streamlining Data Classification Against Enterprise Taxonomies
The Squirro Classifier is designed to address one of the core enterprise pain points. As organizations struggle to extract value from their vast enterprise data repositories, the Squirro platform's ability to fuse retrieval augmented generation (RAG) with knowledge graphs is a key differentiator. The newly added Squirro Classifier streamlines the journey of content from raw data to actionable knowledge.
This cutting-edge feature automatically identifies and categorizes key entities such as people, organizations, locations, and custom entity types within text data, drawing on – and expanding – their existing enterprise taxonomies. By streamlining the otherwise unwieldy data classification process, the Squirro Classifier increases the accuracy and value of GenAI applications that draw on enterprise data.
Enhancing Data Privacy and Integrating External Tools
The platform update also includes new features that address and anticipate common customer needs. PII detection and masking responds to growing data privacy concerns in the age of conversational AI. This strengthened privacy layer de-identifies sensitive information before it is sent to third-party language models, ensuring that data remains secure and compliant with global regulations.
Additionally, Squirro Chat now supports the model context protocol (MCP), an open standard that allows AI applications to securely connect to external tools and data sources. By adopting MCP, Squirro enables its GenAI service to perform more complex tasks by integrating with a wide variety of enterprise systems.
Beyond the major highlights, the new release also introduces a Brave Search Agent for the GenAI service, allowing it to securely access and draw from the public web. A new Search Evaluation Studio provides a framework for measuring the performance of search pipelines, and an updated AWS S3 bucket data loader offers enhanced flexibility for data ingestion from the cloud.
"We are constantly working to address and anticipate the pain points that our customers face in their operations," said Jan Ebner, Head of Product at Squirro. "This release is a direct result of listening to our users and providing them with powerful capabilities that address their specific challenges, whether it's automating data classification, ensuring data privacy, or integrating with their existing toolset."
Bridging Unstructured Data and Structured Knowledge
This process begins with the intelligent annotation of both structured and unstructured content, linking it to existing taxonomies and discovering new entities along the way. This enriched content is then used to automatically build a comprehensive knowledge graph that reveals previously hidden relationships and insights, all of which directly serve to enhance and enrich your enterprise GenAI applications.
For organizations that have spent years meticulously refining their enterprise taxonomies, the Squirro Classifier offers a simple way to increase the potential of their GenAI initiatives and maximize the ROI on those past efforts.
Learn more about the feature updates in the release notes or book a demo to experience them live.
About Squirro
Squirro is a global leader in enterprise-grade generative AI and knowledge graph solutions, specializing in semantic search, insights, and automation. Since its founding in 2012, the company has been at the forefront of AI innovation, empowering organizations with intelligent technology that enhances decision-making and efficiency. Headquartered in Switzerland, Squirro operates across key global markets, with offices in the United States, the UK, and Singapore.
Trusted by industry leaders such as the European Central Bank, the Bank of England, Standard Chartered Bank, Oversea-Chinese Banking Corporation and Henkel, Squirro is committed to delivering cutting-edge AI solutions. Its mission is to augment human intelligence, streamline processes, and unlock actionable insights, enabling enterprises to navigate complexity and harness the full potential of their data.
Contact:
Leo Dufour
Marketing Lead, Squirro:
leo.dufour@squirro.com
