![](/dam/jcr:c7a57b6c-89e1-4b6d-ac59-a1f79bca8543/image-magazin-planet-ai-2.jpg)
At the core of IDA 5.2 is the innovative LLM-based (Large Language Model) data extraction, which proves its efficiency in analyzing unstructured data. "The LLM-based data extraction feature, now in beta with IDA 5.2, is a revolution in document processing," explains Jesper Kleinjohann, CEO of Planet AI. "We use advanced language models, which have become well-known to everyone since the introduction of ChatGPT. However, our technology is specialized in analytical tasks, thus precisely extracting existing data from documents. It identifies relevant information through a deep understanding of context – a capability that goes far beyond conventional approaches, providing our partners and customers with a decisive added value in document processing."
![](/dam/jcr:be9f650b-f601-4caf-84ce-8f8ba838db7b/image-magazin-planet-ai.jpg)
We use Podigree to embed podcast content on our website. Please allow cookies to play the podcast. For more information on the required data and how it is processed, please refer to our Privacy Policy.
Software vendors can significantly enhance the base accuracy for their technologies by using IDA‘s advanced handwriting and machine-print capture, which is now available as a plug-and-play SDK (software development kit). New functions for image binarization and cropping dramatically improve the readability and storage efficiency of documents, while the redaction of data fields ensures the protection and security of sensitive information.
The further development and optimization of IDA 5.2 demonstrate ongoing dedication of Planet AI to research and development, supporting companies on their journey towards digitalization with the highest data quality.