Skip to Main Content
QUL logo

Artificial Intelligence

Artifical Intelligence and the research process

Transparency

Training data – most AI developers will not provide any detailed information about the data used to train their AI models. At best, they may indicate the data came from a variety of public internet sources.

Citations and references – most AI tools will not provide any sources or references for their outputs. There is no way to distinguish between the opinion of a blog writer and a fact found via scientific experimentation published in a peer-reviewed journal.

Algorithms – while the general nature of AI algorithms is well-known, the specifics of how each AI model works is a ‘black box.’ In fact, it may not even be possible for the developers to fully explain why an AI tool produced a certain output.