Glossary

This glossary contains some key terms related to artificial intelligence that may be useful for you to know, especially if you are seeking further information. 

Academic Integrity

#Artificial Intelligence (AI)

#Artificial General Intelligence (AGI) 

#Agentic AI

#Deep learning

#Generative Artificial Intelligence (AI)

#Generative AI misconduct

#Foundational Model

#Large Language Models (LLMs)

#Machine learning

#Neural Networks


Academic Integrity: conducting academic pursuits according to the values of honesty, fairness, trust, and respect, and taking responsibility for one’s academic work and commitments.

Artificial Intelligence (AI): A term coined by McCarthy (1955) originally defined as “the science and engineering of making intelligent machines.” Today, the modern understanding of AI emphasizes methods through which machines can learn and generate outputs such as content, forecasts, or recommendations for human-defined objectives. Google describes AI as “…a field of science concerned with building computers and machines that can reason, learn, and act in such a way that would normally require human intelligence or that involves data whose scale exceeds what humans can analyze.” 

Artificial General Intelligence (AGI) (sometimes General(ized) Artificial Intelligence): A type of artificial intelligence (AI) that matches or surpasses human capabilities across a wide range of cognitive tasks (Heaven, 2023). This is different from narrow AI, which is designed to address a particular set of tasks. Google, OpenAI, and Anthropic, among others, are all working towards development of AGI.  

Agentic AI:  Agentic AI systems are distributed artificial agents that can solve complex problems with minimal human intervention. These agents can plan, predict outcomes, and take actions to achieve goals autonomously. They are also able to recall and use past experiences and knowledge to inform and refine strategies, and control and use external tools to achieve their goals. 

Deep learningDeep learning is a type of machine learning and artificial intelligence (AI) that imitates the way humans gain certain types of knowledge (Gillis, 2023).

Generative Artificial Intelligence (AI): Technology that creates human-like content — including text, images, video and computer code — by identifying patterns in large quantities of training data, and then creating original material that has similar characteristics. Examples include ChatGPT for text and DALL-E for images (Pasick, 2023).  

Generative AI misconductThe misuse or unauthorised use of Generative AI tools 

Foundational Model: An AI model that is trained on broad data at scale, generally using self-supervision, designed for generality of output, and which can be adapted to a wide range of tasks and operations (e.g. et al (2021), The E.U. AI Act), for example GPT-4, Llama, and Mistral.  

Large Language Models (LLMs): Large Language Models are the foundations that underlie text-based generative AI applications such as ChatGPT.  

Machine learning: Is a field of study in artificial intelligence focused on the development of statistical algorithms that can learn from and generalize to unseen (and often unstructured) data, allowing them to perform tasks without explicit instructions. The term was coined in 1959 by Arthur Samuel (Samuel, 1959), who worked for IBM and was a pioneer of artificial intelligence. Machine Learning algorithms search for patterns and relationships in data and can improve their own performance over time (Gillis, 2023). 

Neural NetworksAn artificial neural network is a group of interconnected artificial neurons interacting purposefully. It resembles the human brain in two respects: The knowledge is acquired by the network through a learning process, and interneuron connection strengths known as synaptic weights are used to store the knowledge. Neural networks were first proposed in 1944 by Warren McCullough and Walter Pitts and the term is often used interchangeably with deep learning.