Skip to Main Content

Generative AI: Supporting AI Literacy, Research, and Publishing

Key Questions for Evaluating AI Tools & Outputs

When encountering new information, the AI-literate citizen interrogates the tool, its outputs, how it is represented by the media/field experts, and considers the following questions:

  • What level of transparency exists surrounding the creation of the tool?
  • Has the underlying training data, as well as the methods for obtaining it, been identified?
  • What biases result from the training data and algorithms?
  • What prompts were used to produce the information?
  • What is the purpose of the information output?
  • Is the information output factual?
  • Does the information output contain proprietary content, including that protected by copyright or intellectual property laws?
  • Was AI use disclosed as part of the information creation process?

Applying the ROBOT Test

Librarians Sandy Hervieux and Amanda Wheatley, both of McGill University, created a framework - "The ROBOT Test" for evaluating AI tools (especially as these technologies and their respective outputs are presented by the media).


The ROBOT Test

Being AI Literate does not mean you need to understand the advanced mechanics of AI. It means that you are actively learning about the technologies involved and that you critically approach any texts you read that concern AI, especially news articles. 

You can use the ROBOT Test tool when reading about AI applications to help consider the legitimacy of the technology.



  • How reliable is the information available about the AI technology?
  • If it’s not produced by the party responsible for the AI, what are the author’s credentials? Bias?
  • If it is produced by the party responsible for the AI, how much information are they making available? 
    • Is information only partially available due to trade secrets?
    • How biased is the information that they produce?
  • What is the goal or objective of the use of AI?
  • What is the goal of sharing information about it?
    • To inform?
    • To convince?
    • To find financial support?
  • What could create bias in the AI technology?
  • Are there ethical issues associated with this?
  • Are bias or ethical issues acknowledged?
    • By the source of information?
    • By the party responsible for the AI?
    • By its users?
  • Who is the owner or developer of the AI technology?
  • Who is responsible for it?
    • Is it a private company?
    • The government?
    • A think tank or research group?
  • Who has access to it?
  • Who can use it?
  • Which subtype of AI is it?
  • Is the technology theoretical or applied?
  • What kind of information system does it rely on?
  • Does it rely on human intervention?


Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Hervieux, S. & Wheatley, A. (2020). The ROBOT test [Evaluation tool]. The LibrAIry.

Fact-Checking and Verifying Outputs


ChatGPT and other Generative AI tools have been known to "hallucinate"; that is, they can produce false information and can do so convincingly. For example, GPT3, 3.5, and 4 have been known to fabricate fake citations to works that don't exist. The unsuspecting user then incorporates these fake citations into their own outputs, and unknowingly promotes misinformation. 

To be AI-literate, users must constantly question and fact-check the information that Generative AI tools like ChatGPT produce. While this can be cumbersome, not exercising due diligence can have serious repercussions. 

  • Case 1: Steven A. Schwartz, an attorney who used ChatGPT to create a legal brief as part of a lawsuit. ChatGPT cited court decisions that did not exist. Mr. Schwartz and his law partner, Peter LoDuca, failed to fact-check the brief and submitted it as part of the legal record. The false citations were discovered, and Mr. Schwartz and Mr. LoDuca's professional reputations are at stake. Not only that, the legal ramifications are tremendous, as both are currently being considered for sanctions. 
  • Case 2: Jonathan Turley, a law professor at George Washington University, was listed by ChatGPT as a law professor who had sexually harassed a student on a 2018 trip to Alaska. In addition, ChatGPT cited a Washington Post article to bolster these claims. However, the Washington Post article doesn't exist, Turley has not been on a student trip to Alaska, and claims he's not been accused of sexually harassing a student.

Verifying Outputs

Although there are multiple tools on the market currently advertising the capability to differentiate content that has been produced by Generative AI from that of humans, no single tool can - without human intervention and critical thinking - do so at this time. Additionally, in a submitted (not yet peer-reviewed) paper to the International Journal for Educational Integrity, researchers from multiple European research institutions conducted a study on the efficacy of these tools in detecting AI-generated content and concluded "that the available detection tools are neither accurate nor reliable and have a main bias towards classifying the output as human-written rather than detecting AI-generated text."

That said, the following resources may have utility - in conjunction with critical thinking and information literacy skills, as well as the ROBOT Test - in assisting users with deducing the likelihood of AI involvement.