KNIME, one of the leading open-source data science and AI companies, is unveiling new generative AI (GenAI) capabilities that will empower organizations to securely scale its use and broaden the availability of models. Equipping enterprises with a set of internal controls, KNIME’s release enables the secure use of GenAI according to an enterprise’s unique risk assessment and internal policies.
The risk of GenAI leaking sensitive data is a major obstacle toward its proprietary success, according to KNIME. Driven by this risk factor, KNIME’s latest release allows organizations to take advantage of the latest and greatest AI technology while offering extensive controls, including dedicated model routing, personally identifiable information (PII) anonymization, access management, and more.
“Governing data science processes is—or at least should have been—on people’s radars for years. The risks presented by GenAI are helping people realize that,” said Michael Berthold, founder and CEO of KNIME. “KNIME was built from the beginning to help with the continuous, safe deployment of data science processes in mind, so it was straightforward for us to add a customizable set of controls around GenAI.”
“KNIME provides a single environment for data scientists that balances innovation and control by giving data workers access to as much new GenAI functionality as possible while allowing them to control risk by putting the right governance mechanisms in place,” continued Berthold.
KNIME users can now connect to the latest AI models from commercial vendors—such as OpenAI, Azure OpenAI Service, and Databricks—as well as local large language models—such as Llama 3—from within the KNIME platform. Users can also connect to local and remote Hugging Face Text Embedding Inference servers, granting access to a wide variety of open source embedding models that make features like semantic search and feature extraction possible. Additionally, KNIME users can connect to protected Hugging Face Inference Endpoints to rapidly deploy GenAI models for experimentation.
This release also introduces a GenAI Gateway to KNIME, enabling IT to configure which GenAI providers and nodes can be used by team members for enhanced governance and security. Integrations with Presidio and Giskard also further security, serving to protect PII against leakage to external LLM providers as well as evaluating machine learning (ML) workflows for robustness and bias.
To learn more about KNIME’s latest release, please visit https://www.knime.com/.