Generative AI can be a powerful tool, but it is just a tool. To use it to your benefit, you need to understand what it’s good for and what it’s not. It can be helpful, and it can also work to your detriment.
As a tool, you should remember that AI is only there to suggest, and never make decisions or do any of the creative or substantive work for you. It is great for suggesting directions, spell checking and refining your edits, but it shouldn’t write your paper for you, or do your research for you either.
For the research process taught in INFO 1010 and other similar classes, Generative AI can be helpful for some of the minor parts of the process. An example of that is the brainstorming and background research part of the Planning stage of the process. It can help you generate ideas to consider, but it can’t make that decision for you. You get to decide what you write about and how you want to write it.
When using AI to brainstorm, you can take a broad topic and ask the AI tool to suggest some academic topic options for you to consider. It might have some background information on the topics it suggests, but since AI is not as up to date as other information sources such as Wikipedia, it definitely shouldn’t be the only source you use.
Another thing AI is helpful for is figuring out the academic keywords for a topic. Often we can only think of and explain our topics in the vernacular, or natural language. AI can be a helpful tool for brainstorming those keywords and translating natural language into academic keywords and language.
Hallucinations: You should also always fact check AI, because it is prone to “hallucinate” sources and quotes. If AI doesn’t know the answer, it will make stuff up. We call these hallucinations. In research, AI will fabricate sources that don’t exist, but look genuine because they are based on other real sources, etc.
Don’t use Generative AI to create your citations. This is another instance where the AI will hallucinate and create fake sources. It’s better to use a citation generator and then proofread, because even citation generators are liable to make mistakes.
Another downside of using AI is that it will only summarize the known knowledge on any topic. It will be outdated, since the language models are not trained on the very latest resources, and it will not have anything new or interesting. AI will give you very basic ideas that have been used a lot without anything necessarily interesting and definitely not groundbreaking. Don’t limit yourself to just what it suggests. It’s a starting point, not the ending point. As always, you should be making the decisions and exploring further than AI will.
Generative AI is also only able to use what is freely available on the internet, where so much of the best resources are behind a paywall, so the quality of what Gen AI is able to access isn't high enough for academic writing, or it will just use what little is available. This will make anything it creates at a subpar standard, especially if it’s hallucinating sources, and will not be doing you any favors if you rely on it for this type of work.
When using Generative AI for brainstorming, be sure to be very specific in what you are asking, such as asking for academic or scholarly topic options, to make sure you are considering the right kind of topic for your research.
Generative AI can also help you brainstorm keywords, helping you translate your words from the vernacular or everyday language to academic wording. Be sure to specify that you need academic or scholarly, etc., so it knows what you are looking for, and that vernacular is less helpful.
Once again, it’s up to you to sort through the options given and see if they are worth using. Never just take anything provided by a Generative AI at face value, but always sort, screen, and fact check the results, and then use the results to make your work easier and better. Never let Generative AI replace your own thinking. You have to make the decisions.
