Interest in the use of AI in defence has massively increased thanks partly to the widespread use of large-language models such as ChatGPT and partly to the investment defence is making in AI in the wake of Russia's invasion of Ukraine.

Alas, with this interest comes a massive increase in the hype, misunderstandings and unrealistic expectations created by those who talk about AI without having any practical experience of developing and implementing AI in defence. This poses the risk of huge disappointments and a resulting backlash that could hold back the military advantage we get from AI. What can defence do to minimise the risks of falling into the cycle of overpromise, overspend, underdeliver?

This book offers one solution; to learn the lessons from those who have actually developed and implemented AI into defence, from initial concept to use by defence organisations. It covers the practicalities of defining what AI actually is, understanding how defence deals with trust and ethics, the steps that lead to the utilisation of AI for a range of different use cases, and how to see past the potential smoke and mirrors when testing AI.