AI can help with academic research but doesn’t always get it right – here are some tips to check the information you receive
AI is quickly changing. These tips from university librarians can help people keep pace with the shifts as they approach their research.
Nearly one quarter of adults in the United States use artificial intelligence chatbots daily, according to a 2026 Pew Research Center study. However, a recent study revealed that 48% of the information provided by these popular free chatbots in response to questions about current events is incorrect. Librarians caution that students using AI chatbots for academic research should not assume the information is accurate.
To ensure the accuracy of information obtained from AI tools, librarians recommend following four steps:
1. Choose the appropriate AI tool: Different AI tools, such as ExLibris' Primo Research Assistant, Research Solutions' Scite, Anthropic's Claude, or OpenAI's ChatGPT, are trained using various datasets and algorithms. They should not be used interchangeably as some tools may be better suited for specific tasks or generate varying results.
Students can also explore library-specific AI tools like LeapSpace or Concensus that draw their data from peer-reviewed scholarly research. Evaluating reviews of these tools can help students select the most suitable option.
2. Ask follow-up questions: When using AI tools, it is essential to ask questions to better understand their responses. Even if an AI-generated answer appears correct, further investigation is crucial. Students should inquire about the source of the information, the tool's confidence in its accuracy, and any potential gaps in the response. They should also examine how the AI arrived at its conclusion and ensure the prompt is neutral in tone.
3. Trace claims to original sources: Students should verify the sources cited by the AI tool to ensure the accuracy of the information. They can follow the SIFT process used for evaluating websites, asking the AI to provide sources for its claims. By examining each source, students can determine if the AI used the information appropriately within context and investigate the authors' expertise in the topic.
If the AI does not cite sources or their credibility is questionable, students should conduct additional research using reliable libraries or databases.
4. Recognize less visible issues: AI tools may not always highlight important perspectives or present unbiased information. They often focus on freely available, English-language content written on the internet and may lack expert knowledge from other sources. Students should be aware that AI tools may miss crucial viewpoints and seek out these missing perspectives when needed. Lastly, when quoting or paraphrasing AI-generated text in their work, students must cite the AI tool as they would any other source.
Written by urgent.news from The Conversation's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
