OneSearch has an AI component called "Research Assistant."
First, log into OneSearch. The "Research Assistant" link is found on the blue "ribbon" at the top of the page. You will be prompted to enter your BTC Network Login credentials (e.g. smith4321 and the password you created). Don’t remember your password? Reset your password using the Microsoft Password Reset tool.
To start go to OneSearch and click at the top on "Research Assistant." 
This page shows up next. Enter the search query and select filters by clicking on the grey icon with lines and circles.
Select TYPE to narrow the search results. Select "Peer Reviewed" for scholarly articles. 
Select DATE to chose the date range. For current research, choose "Last 5 years."
Select AVAILABLE ONLINE to see only articles you can view and read right now.
Type in your research question. Your choices will be highlighted.
An overview will be displayed with the sources noted. You can then choose to view/read the article best suited to your research question.
Click on the article you want to find more information about.
To access article details, click on the title of the article you selected to open its details. Then, click on the link that says “More Details and Request Options” to view more information about the article. Note: There may be links/PDFs available for the article; not all relevant articles are available full text.
Evaluate the Article Using the CRAAP Test: Use the CRAAP test (Currency, Relevancy, Authority, Accuracy, and Point of View) to check if the article is trustworthy and accurate. This will help you decide if the article is a reliable source for your research.
AI Insights - Beta reduces the time users spend searching for relevant full-text articles supporting their research. This feature is accessible across a vast and continually expanding array of EBSCO full-text databases and EBSCO Discovery Service (EDS) via multiple interfaces.
AI Insights offer concise summaries, presenting 2-5 key points for each document. This consistent format, applied across multiple databases and deep backfiles of licensed content, introduces a new level of uniformity, facilitating rapid review.
How AI Insights are generated
AI Insights summaries are generated by prompting a Large Language Model to summarize insights from the specific article the user selected AI Insights for. The AI Insight prompt uses a method called Retrieval Augmented Generation (RAG) to reduce hallucinations. EBSCO also reviews a representative sample of AI Insights with Subject Matter Expert (SME) Human-in-the-Loop (HITL) for biases, tone, accuracy, and timeliness of Insights as a quality and responsible AI metric.