The open source company KNIME , specializing in data analysis and the secure use of artificial intelligence (AI), has equipped the KNIME Data Analytics Platform with an AI assistant – K-AI (KNIME AI). From now on, all users can also create and optimize sophisticated data workflows with AI. K-AI provides answers to specific questions, makes recommendations and designs or expands entire data workflows based on user requests. This not only saves time. With K-AI, users also get maximum transparency and control over every step the AI takes.
The KNIME platform was developed to give everyone the opportunity to carry out complex work with data – regardless of their technical knowledge level. To do this, KNIME uses visual workflows that can be set up and adapted even without programming knowledge. The secure integration of AI is a significant advance in the automation of data work, enabling users of all experience levels to gain better insights from their data more quickly.
In contrast to conventional AI tools that only generate scripts, KNIME’s intuitive visual workflows help data workers to immediately understand the contribution of AI. On the one hand, the visual workflow serves as documentation of each step in the process of data manipulation or analysis. On the other hand, a narrative text generated by K-AI complements the AI’s decisions and methods, making them easy to understand.
Working through visual workflows marks a paradigm shift for the use of AI in data work: away from opaque, code-heavy environments and towards transparent, documented processes that everyone can understand.
This makes K-AI a reliable partner in data work
Many companies have set themselves the goal of becoming more “data-driven” – i.e. acting on the basis of data. However, they often face considerable challenges in doing so – whether it is working efficiently with data, having too few data scientists or because lengthy data preparation processes make timely insights based on data impossible.
Also Read: Qlik Eases Cloud Migration for Seamless AI Adoption
K-AI helps everyone in the company to train more quickly so that they can carry out data work independently, while at the same time relieving the burden on internal experts. In addition, K-AI significantly increases the speed with which data can be prepared and cleaned for analysis – a process that is very time-consuming to do manually and involves a lot of routine work.
K-AI offers two modes:
- Q&A mode : Users can use K-AI to ask technical or theoretical questions about KNIME. K-AI provides answers that explain to users how certain nodes work within a workflow or how they can optimize existing workflows.
- Build mode : In interactive design, K-AI works directly with users. Based on user instructions, K-AI adds, connects, configures nodes and continuously offers suggestions for the next step. Whether it’s creating simple workflows with a single prompt, formatting visualizations or generating embedded Python or R scripts, K-AI explains each step, what and why changes were made.
KNIME’s AI assistant not only helps you create completely new workflows. In Build mode, existing workflows can be easily extended, while Q&A mode offers targeted recommendations for extensions, workflows and nodes from the KNIME Community Hub. This makes it easier than ever for users to access proven contributions and templates from the KNIME community and find solutions without having to start from scratch.
“KNIME’s mission has always been to empower everyone to understand their data, regardless of their technical background,” explains Michael Berthold, CEO of KNIME. “With K-AI, we are taking a significant step forward by helping data workers build workflows faster while learning more. By combining AI support with KNIME’s open source platform, we are not only accelerating the way users work with data, but also enabling them to learn and grow over time. Initial feedback has been overwhelmingly positive, and we are excited to see how K-AI will help users unlock new possibilities with their data.”
K-AI was first released as an experimental extension in early 2024. During the experimental phase, both the K-AI Q&A mode and the Build mode were tested and refined. Now, K-AI is being released for the first time in KNIME for all users as part of the broader KNIME Analytics Platform version 5.4 , which includes numerous improvements to LLM connectivity, LLM evaluation capabilities, and many additional enhancements:
- Improved AI extension with support for Databricks LLMs and GPT4All models on IBM Granite and Microsoft Phi 3.5 in addition to existing support for OpenAI, Azure, Hugging Face and Ollama models.
- The ability to detect vulnerabilities such as distortions, hallucinations, data leaks and more in AI-powered workflows using Giskard LLM and Giskard RAG nodes.
- Several connectivity improvements for Tableau, Box and Databricks in addition to KNIME’s 300+ existing data source integrations.
Starting with KNIME Analytics Platform 5.4, K-AI is included for all users.
Source: Businesswire