Defence Artificial Intelligence Centre publishes Defence AI Playbook

The Defence Artificial Intelligence Centre publishes its ‘AI Playbook’ demonstrating the breadth of opportunities that AI presents.

Following the 2022 Defence Artificial Intelligence Strategy, in January 2024 the Defence Artificial Intelligence Centre (DAIC) published its Defence AI Playbook with the purpose of demonstrating the breadth of opportunities ‘from strategic advantage on operations to efficiency in our business processes’ that AI presents to defence.

Established by the 2021 Integrated Review, the DAIC is the organisation within the Ministry of Defence tasked with leading the transformation of UK Defence into an ‘AI-ready’ organisation, and building the necessary ‘ecosystem’ of businesses, academics and others required in order to do this.

The Playbook identifies what it calls the six AI ‘problem spaces’ for defence with practical examples of how each could be deployed operationally:

  1. Recognise – detect subjects of interest by matching patterns in the vast volumes of sensor data MOD collects

To enhance recognition capabilities including both visual and oral, through satellite imagery analysis and Radio Frequency surveillance.

  1. Comprehend - derive insight from (often large) unstructured and semi-structured datasets

To assist analysts with extracting information from documents, and using natural language capabilities to rationalise and simplify existing policies.

  1. Predict – Anticipate likely outcomes of future events based on historic data

To use historic data to predict demands such as vehicle failure reports to anticipate future part requirements.

  1. Simulate – explore scenarios and analyse data to inform planning, including alternative courses of action

To use Machine Learning and National Language Processing to automate and streamline planning processes ‘such as automating the cross correlation of terrain… against vehicle characteristics and route planning’.

  1. Generate – create new and original content from the patterns and structures within existing data

To understand how Large Language Models (LLMs) can be used by Defence safely and securely.

  1. Decide – create autonomous or automated behaviours by selecting the actions to achieve a goal

To manage uncrewed, autonomous platforms such as reconnaissance drones, mine hunters and last-mile resupply.

To read the full report click here.

Jeremy Wimble

Programme Manager, techUK

Jeremy is a programme manager at techUK and is responsible for the delivery of the defence programme’s activities through the Defence and Security SME Forum, Defence Commercial Business Forum, Defence Research and Technology Forum, and Information Superiority Forum.

Prior to joining techUK, from 2016-2024 Jeremy was International Security Programme Manager at the Royal United Services Institute (RUSI). In this role he coordinated the team’s research and impact activities for funders including the Foreign, Commonwealth and Development Office and US Department of Defense. He also worked on business development and strategy.

Jeremy has a MA in International Relations from the University of Birmingham and a BA (Hons) in Politics & Social Policy from Swansea University.

[email protected]

Read lessmore