Skip to Main Content
Today's Hours:
Menu

Artificial Intelligence (AI) for Library Research

AI tools can be incredibly powerful, but they also have their limitations. The information generated by AI can be inaccurate or misleading; AI can fabricate research, generate biased answers, or not acknowledge authors for their work. This may ultimately weaken rather than enhance a student’s understanding of a topic. Students should be aware of errors that may arise and should double-check any output retrieved from AI tools, including citations to verify accuracy.

Limitations

AI systems can display various biases in their responses due to weaknesses in the data on which the AI systems are trained:

  • Confirmation bias: AI can unintentionally reinforce existing beliefs or opinions present in the training data, resulting in biased outputs that align with specific perspectives or ideologies
  • Content bias: Al models may show biases in the types of content they generate, favoring certain topics, themes, or perspectives over others
  • Gender bias: AI models can replicate and perpetuate societal gender biases, including stereotypes and gendered language
  • Racial and ethnic bias: AI models may generate content that reinforces stereotypes or unequal treatment of differnet racial and ethnic groups
  • Cultural bias: Models trained within specific cultural contexts might produce content biased towards or excluding other cultures, leading to misrepresentation or lack of representation

Hallucinations occur when AI generates incorrect or nonsensical information. 

  • Misinformation/Disinformation: Hallucinations can contribute to the spread of false or inaccurate information, leading to negative consequences for individuals and society.
  • Decision Making Errors: If AI systems are used to support decision making, hallucinations can lead to incorrect and harmful outcomes.

Plagiarism occurs when AI systems generate content that is substantially similar to existing copyrighted material without proper permission or proper attribution. 

  • Legal Implications: Plagiarism can lead to copyright infringement lawsuits if someone uses copyrighted materials without the permission or acknowledgment of the owner
  • Academic Integrity: Using someone else's work without proper attribution can result in a violation of UMW's Honor System.
  • Reputation and Credibility: Plagiarism undermines the trust and credibility of individuals or organizations who inappropriately use someone else's property without their permission or without attribution

Using an AI tool to conduct your research might be helpful, but they do not provide access to the overwhelming amount of full text books and articles that are available through the library.

  • Full Text - Search the library for books and articles that are not available through an AI search engine