Whenever any new technology emerges, it inevitably generates a host of new words and phrases to help describe its function and components. Artificial intelligence (AI) is no different. AI is a revolutionary yet poorly-understood technology – capable of changing the world, but often greeted with trepidation or outright fear by the media and members of the public alike. This glossary of AI terms will help to disambiguate some of the most common words, phrases, and acronyms around artificial intelligence – making the technology easier to appreciate and understand.
AI (Artificial intelligence): Artificial intelligence is commonly abbreviated to AI. The phrase describes computer systems designed to perform tasks that would normally require human intelligence to complete.
AGI (Artificial general intelligence): Artificial General Intelligence describes an AI system that has achieved human-level intelligence and abilities to learn and improve. While many AI systems can perform tasks quicker or more accurately than humans, none has yet achieved human-level intelligence. Most experts expect AGI to be reached within the next 35 years.
Algorithm: An algorithm is a set of instructions that tells an AI system what to do. It’s the programming that determines how AI tools make decisions.
ANI (Artificial narrow Intelligence): Artificial narrow intelligence is a term that describes an AI system designed to perform a single, specific task. All current AI applications are forms of narrow AI.
ASI (Artificial superintelligence): Artificial superintelligence describes an AI that is more intelligent than the smartest human in every measurable way. Some experts believe ASI will never be achieved, while others expect it to happen on a similar timeline to AGI.
Deep Learning: Deep learning is a subset of machine learning. It uses neural networks to ‘learn’ by recognizing patterns in large volumes of data.
GenAI (Generative AI): Generative AI is a type of artificial intelligence used to ‘create’ content: text, images, video, music, and more – although the extent of GenAI creativity remains controversial.
LLM (Large Language Model): LLMs are machine learning models trained on large volumes of text, allowing them to ‘understand’ human speech. They’re capable of analyzing text and creating written content.
ML (Machine learning): Machine learning is a subset of artificial intelligence. ML programs can ‘learn’ from data – allowing them to improve through repetition and carry out tasks without specific instruction.
Neural Networks: Neural networks are a type of machine learning. They’re an attempt to recreate the way human beings process data through a series of interconnected nodes, or ‘neurons’.
NLP (Natural Language Processing): Natural language processing is a way for machine learning systems to understand the way humans communicate. It allows users to extract insights from large volumes of information quickly and accurately.
Supervised Learning: Supervised learning is a means of building machine learning models with labeled data sets. This approach is very useful in applications with well-defined goals, including life science.
Training Data: Training data refers to the data set on which an AI program has been trained. For example, an LLM might know about literature or journalism because it’s been trained using books, newspapers, and academic texts.
Transformer: A transformer is a type of deep learning architecture composed of neural networks. They’re used to understand context and relationships within the data.
Unsupervised Learning: Unsupervised learning is the opposite of supervised learning: namely, machine learning models built with unlabelled data sets. These models look for patterns without human input – making them useful for uncovering insights you didn’t know were there.
Artificial intelligence is a complex discipline that’s evolving all the time – but it’s already delivering results in life science. You can read more about Within3’s AI, our approach, our proprietary technologies, and how it can benefit your organization here, but if you’re ready to experience it for yourself, you can also schedule a demo here.
