Skip to Main Content

Generative AI for Learning and Research

How to choose your GenAI tool?

There is a wide range of GenAI tools available, from elementary level to the expert grade, from generation of text to audio and video. With so many choices, how should you choose the tool that best suits you? Below are some suggestions for you to consider before deciding:

  1. Understand your needs
    First, it is important to understand your needs—what you would like the GenAI tool to help you with. Are you looking for an image for your presentation? Or would you like the tool to help you with constructing your search topic and summarizing information? Once you have identified your need and purpose in using the tool, you can narrow down the scope to searching for relevant resources.
     
  2. Accessibility and user experience
    Accessibility and user-friendliness of the tool would affect the perspective of user in assessing whether it is suitable or not. Most common consideration would be the pricing of tool, of whether it is free or additional charge is required to unlock a particular feature or not. For some tools, even they are free, availability of access is based on region of the user – like OpenAI and Google Gemini which do not support usage in Hong Kong. User interface and experience, such as the ease of use, interface design and clearness of instructions, would also influence your decision on whether it suits you or not.
     
  3. Accuracy and reliability
    Accuracy and reliability are undoubtedly the vital elements that we should all consider. This would involve the sources and training data that the tool is using, like whether the data set is up to date. Although some tools would also cite the sources of information retrieved, you are encouraged to examine and evaluate the outputs and have them verified before actual application to your assignment, to ensure trustworthiness of the generated outputs.
     
  4. Ethical considerations
    This involves the privacy of data handling, privacy policies and potential bias that the training model may result in. You are encouraged to read carefully their terms and conditions before actual usage of the tool, or to look for tools with open information.

 

References: