Table 1.

Glossary

Term	Definition
Artificial intelligence (AI)	The science of developing computer systems that can perform complex tasks approximating human cognitive performance.
Natural language processing (NLP)	A branch of information science that seeks to enable computers to interpret and manipulate human text.
Large language model (LLM)	A type of NLP model comprising large neural networks trained over large amounts of text, usually to produce an output of continuations of text from corresponding prompts, known as next word prediction.*
Multimodal LLM	LLMs with the capacity to integrate input from various data types, including text, speech and/or visual sources.
Next word prediction	The NLP task of predicting the next word in a sequence of text given context and model parameters.
Parameter	A parameter within an AI algorithm is a variable that is tuned iteratively/automatically to optimise the intended outcome of the algorithm. Parameters may be at the model level to optimise tuning (hyperparameters) or ‘weights’ within the model linking layer to layer (parameters).
LLM-linked chatbot	A program that permits users to interact with an algorithm (such as an LLM) designed to respond to user prompts.†
Chatbot Assessment Study	Any research study assessing the performance of chatbots in summarising health evidence and/or providing clinical advice.
Chat instance	An interface in a computing device through which communication takes place between a chatbot and its user through text with only one prompt.
Chat session	An interface in a computing device through which communication takes place between a chatbot and its user through text with more than one prompt.
Query	The act of communicating with an LLM by inputting a prompt into the chatbot which might be a question, comment or phrase to elicit specific desired outputs from an LLM. For example, one might input a prompt asking the LLM to summarise the evidence supporting the use of a given intervention.
Check query	Following formal query completion and performance evaluation, the act of repeating the initial query to ensure that chatbot outputs are consistent in summarising the same evidence and providing the same clinical advice.
Prompt	Text input by a user into the chatbot for the purpose of communicating with the LLM.
Prompt engineering	An iterative testing phase where various pieces of text are inputted into a chatbot to achieve an output, informing the development of study prompts.
Delphi study	A structured research method applied to answer a research question through the establishment of consensus across respondents.

*Generally speaking, ‘next word’ prediction is one basic ‘pre-training’ objective, but LLMs often undergo a subsequent round of ‘supervision’ in which they are guided by human feedback.

†Chatbots are not necessarily built atop LLMs, but the modern tools that have captured public imagination are.