Research Guides: Integrity with Artificial Intelligence: A Compass for the Underground Way

A Cautionary Tale

Two major problems with AI-generated content are confabulation and bias.

Confabulation is the creation of false content without the intent to deceive. AI tools, especially chatbots, are prone to misconstruing or even making up information because the tool is designed to generate quick, unique content (without human analysis).

If an AI chatbot is asked to summarize the latest dementia research, it may comb through hundreds of articles but misconstrue the information without having a big picture, experienced understanding of dementia research. AI can also misinterpret the complexity of references and incorrectly cite sources. It may create plausible-sounding information and sources but without the ability to incorporate human synthesis and critical thinking, it cannot verify the accuracy of what it is composing and citing.

Algorithmic bias can occur if the datasets AI tool rely on contain bias themselves. This is often the case because the underground terrain of AI consists of human beings choosing what data and information can be used, often based on commercial or other interests. Common biases that can occur in the information AI tools utilize can include biases against race, class, religion (Christian history and thought) and political beliefs.

Because of both confabulation and the strong potential for bias, students MUST evaluate whether the information and citations the AI tool generates are good sources of information. See how to evaluate AI content below.

Evaluating AI Content

Accuracy: AI-generated content is not always accurate. It may contain errors, false claims, or plausible-sounding content that is invented and false (confabulations). AI tools may be limited by the dataset available to them, which may not include the latest information. Consider that subscription-based resources, such as library databases, and the guidance provided by experienced library research professionals, along with knowledge of your professors, may offer more authoritative sources.

Fact checking is very important when using AI tools like ChatGPT!

Bias: AI-generated content may contain biases introduced through the algorithmic design, data collection, data labeling, or model training processes. It’s important to evaluate AI-generated content for common types of bias, such as racial bias, gender bias, class bias, disability bias, religious bias, and political bias.

Currency: AI-generated content may contain information that is outdated. This may result from access to old or limited datasets. For instance, some free GPTs only have access to a snapshot of the internet from several years ago, so they are unable to generate content that draws from the latest information.

Comprehensiveness: AI content may be selective as it depends on the algorithm which it uses to create the responses, and although it accesses a huge amount of information found on the internet, it may not be able to access subscription-based information that is secured behind firewalls. Content may also lack depth, be vague rather than specific, and it may be full of clichés, repetitions, and even contradictions.

Copyright: AI-generated content may infringe on copyrighted material. This is because AI tools often draw upon datasets that include copyrighted material, so new content created from these datasets may inadvertently violate copyright provisions.

Citations: AI-generated content may often lack citations to back up the information provided. If citations are included, they may be invented or misconstrued. Always double-check that citations are real and that the information provided is accurate.