A search function for searches in natural language (Natural Language Search - NLS) is available, in which queries can be made in everyday language instead of search terms. The search mode must be activated for the search:

In the advanced search, under Search options → Search modes → Search in natural language is selected or in the simple search the slider is set to "Search in natural language".

EBSCO uses Natural Language Processing (NLP) to help the database better understand user intent and contextual clues in the queries.

After submitting the query, the "Show refined query" function can be selected above the hit list. A Boolean version of the natural language query is now displayed. You can now see how the natural language query is interpreted by the system and can check whether the AI has understood the intention correctly.

The AI Insights feature creates summaries from full-text articles via the "Generate AI Insights" button. However, this is still a beta version and is only used for licensed databases, but not for e-books.

The summaries are generated using a Large Language Model (LLM). According to EBSCO, the use of the Retrieval Augmented Generation (RAG) method is intended to reduce hallucinations.

The "Generate AI Insights" button is displayed in the hit list if the full text is available. If no full text is available or the content is too short, the button is not displayed. AI Insights does not replace the abstract of the document, which can be found in the details or directly at the beginning of the article after calling up the hit. It is intended as a supplement for a simplified reproduction of the content.

AI Insights are available in the language of the user interface, regardless of the full text language.

According to EBSCO, the model is not trained using customer data, user behavior or other private (or non-private) information. EBSCO only sends the full-text source document to the model with a request to summarize the content.