It is hard to know what to make of claims like “LLMs are intelligent”, “we have reached AGI”, or “AI’s outputs are biased” without a grasp of the definitions of the terms ‘intelligent’, ‘AGI’, and ‘bias’. And yet, many do just this. Interdisciplinary debate would be easier and more fruitful if a common set of working definitions were used. This glossary aims to contribute to that end.

[T]he glossary is underpinned by a rigorous editorial process. To ensure accuracy:

• Entries are written by suitably qualified academics and practitioners with relevant expertise.
• All entries are peer-reviewed by The Glossary’s editorial team, composed of internationally recognised researchers with earlier career academics.
• Entries are often also reviewed by a wider network calling on leaders in specific fields.