How to use AI for research: Practical tips for responsible use

Hey there, friend!

By this point, you’ve probably already heard all about how GenAI is being misused in the workplace and at school: “personal” essays composed entirely by ChatGPT, AI-assisted reports with non-existent citations, legal arguments supported by hallucinated precedent, and on and on. Browse your feed on any given day, and there’s likely no shortage of GenAI missteps to find, causing many to wonder whether it really is possible to incorporate this technology responsibly into the research process at all. 

Long-time Career Chat readers won’t be surprised to find that our answer is a resounding YES!*, strongly tempered by a bold asterisk that notes GenAI is best used to support work and critical thinking rather than replace it. We’ve talked about how AI is changing education, covered tips on using it for data analysis, and explored practical prompting tips. Today, we’re adding to that discussion by covering some of the ways you can use GenAI to boost your research process (responsibly). 

Whether you’re a student, working professional, or educator, there’s likely something for you here. So, without further ado, let’s get into it. 

But, first: Do you want to build your own GenAI assistant to help with your research? Enroll in Vanderbilt’s Generative AI Assistants Specialization to learn how to tailor a GPT-style language model to align with your world–be it legal, logistical, or scientific. 

🔎How to use GenAI for research (responsibly) 

Research is a relatively complex process. It involves not only gathering information, but also formulating research questions and hypotheses, analyzing data, verifying sources, reviewing literature, and designing research approaches and experiments (among other things). In turn, research is an undertaking that scales in complexity in proportion to the questions you ask and seek to answer. 

You can’t quite rely on LLMs to independently manage all your research tasks. Fortunately, regardless of what kind of research you’re conducting, there are many ways you can leverage GenAI to streamline and support your workflow. Here are some ways you can– responsibly–use generative AI to boost your research process: 

🧠Brainstorm and refine research questions: Due to their conversational abilities, LLMs like Microsoft Copilot and Google Gemini can be helpful sounding boards to bounce ideas off of when you’re first formulating your research questions or designing an experiment. 

📑Surface scholarly articles: Some GenAI-powered research tools, such as Connected Papers or Research Rabbit, can map out relationships between papers, making it easier to find articles related to your research topic.  

🧑🏫Clarify tricky concepts: Unsure if you fully understand a certain concept you’ve encountered in your research? Try asking ChatGPT or Coursera Coach to clarify it for you.

📈Cursory data analysis: You can use GenAI to conduct a quick, cursory analysis of your data to get a general idea of what it’s revealing. This may be particularly useful when performing a sentiment analysis on unstructured data like social media posts or product reviews. 

🖍️Constructive criticism: Receiving feedback is critical to the research process. When you lack a human to review your work, though, you might consider interfacing with an LLM to identify any counterarguments, topic gaps, or opportunities for improvement. 

⚠️PSA: Verify, verify, verify!⚠️

Generative AI is best when it supports your workflow rather than replaces it. This is particularly true when you’re conducting research, because while GenAI can augment many of the steps throughout the process, it can’t actually do them reliably. In fact, in some instances, GenAI may simply make up information, events, or sources that don’t even exist. Plus, in some cases, it may actually confirm certain biases rather than rigorously challenging them. 

Put simply: While there’s a place for GenAI in the research process, it’s not a replacement for critical thought, so make sure to verify all of its claims thoroughly before accepting them as fact. 

✨Boost your productivity with GenAI

Generative AI is transforming how we work. Build the skills you need to support your research workflow and boost your productivity with one of these programs on Coursera: 

For an overview of GenAI, try Google’s AI Essentials Specialization. Learn how to use GenAI to help you develop your ideas, write prompts to get the outputs you want, and leverage AI responsibly.

To build and train personal assistants with GenAI, enroll in Vanderbilt University’s Generative AI Assistants Specialization. Explore how to tailor GenAI to intuitively align with your world, parse documents, and engage in natural dialogue–all while rigorously testing to assure precision and reliability.  

To build GenAI skills for data analytics, try IBM’s Generative AI for Data Analysts Specialization. Gain knowledge of foundational prompt engineering concepts, use GenAI models to draw data insights, prepare data, and create visualizations, and explore the ethical considerations and challenges of using GenAI for data analysis. 

 And that’s all we’ve got for this week, friends! Before diving back into your literature review, consider dropping a comment to let us know how you use GenAI to boost your research process. Until next time! 

Have a career question you’d like us to answer next? Share it below. Still reading? Okay. Well, we’re glad you’re so thorough. Check out this wildcard course, doggonit!