What can LLMs do for you?

Create, Generate, Communicate

generate a variety of text formats such as emails, but also more creative text such as poems, music and letters. They can also be used for coding assisting developers in building applications and experiment in different programming languages.

  • write an email in a formal tone. You must check the tone and content before sending!
  • create a brainstorm for a project and provide some inspiration.
  • analyse qualitative data efficiently, identifying patterns and themes in large datasets faster than traditional methods.
  • draft research proposals and plans and recommend sources.
  • create a lesson plan based on the aims and objectives of your course.
  • create a chatbot to answer basic student questions so that answering to emails could be limited to more complex questions.

This picture was created by Canva Free AI Image Generator

Summarize

summarise information from long articles, news stories, research reports and other sources. Many LLMs have the ability to explore pieces of text and provide commentary.

  • proof-read a piece of writing for spelling mistakes or clarify concepts you find difficult to understand.
  • summarise pieces of text for you and provide a list with the most important bits. You must be careful as these are not always entirely accurate.
  • summarise extensive research findings, clinical notes, and literature reviews.
  • summarise a chapter from a book which you could refine and use in your teaching.

This picture was created by Canva Free AI Image Generator

Contribute to educational content

by educators to help create a course structure and content including assignment and exam questions. Chatbots creation can be used to provide personalised student experience, answer student questions and provide feedback and support.

  • give you feedback on a piece of writing assignment.
  • help with the preparation of lecture material, assignments and exams
  • propose innovative methods to engage students, including interactive activities and discussions

This picture was created by Canva Free AI Image Generator

Never
  • use LLMs to create content which you present as your own work. This is plagiarism and will lead to disciplinary action.
  • rely on the content without reading the text yourself. LLMs make mistakes and suffer from AI hallucinations.
  • share private or sensitive information of yours or other people with a LLM. Uploading someone else’s work on an AI program violates their intellectual property

This picture was created by Canva Free AI Image Generator

LLMs cannot...

Large Language Models (LLM)

They do not think like humans

AI language models, don't "think" in the human sense. Instead, they process information and generate responses based on patterns and data they have been trained on. When using LLMs you should remember that they have no self-awareness, emotions, and consciousness. You should also be aware that LLMs lack of original thought as they have no personal experiences. They can generate creative content by combining existing ideas but in novel ways. Lastly and most importantly you must remember that LLMs lack ethical and emotional judgment.
Large Language Models (LLM)

They are not always accurate

Who is the author of Gitikiti?

ChatGPT replies: 'The author of "Gitikiti" is Michael Aridi'. Don't bother searching it....There is no Gitikiti! The reliability of the information provided by LLMs depends on the accuracy of the data they are trained on. If they are trained with false information, they will produce false information in response to user queries. Moreover, ambiguous or poorly defined queries can result in responses that don't fully address the user's needs or intentions. LLMs can have hallucinations which refers to generating information or responses that are not based on actual data or facts.
Large Language Models (LLM)

They are not fully unbiased

The training data may contain biases that can affect the information and perspectives that LLMs provide. Since LLMs are trained on internet data, they can sometimes reflect these biases in their output. For example, LLMs may incorrectly associate certain traits, professions, or skills with specific genders or ethnicities.

Last Updated on 4 March, 2026