AI also has significant potential for positive uses in the defence industry. The technology can analyse thousands of patterns in what has happened previously to predict what could happen in the future.

The secret behind increasing AI's capability to predict future human behaviours is that people are creatures of habit and routine. The ability to make predictions for events that haven't happened yet could prove invaluable for defence and security forces.

Massive Analytic’s AP platform is capable of combining data from different sources to determine patterns. For example, it can combine image and video footage with real-time activity on social media to generate a clearer picture of the most likely future scenarios.

Frangou explains how his company has made efforts to stay away from the area of active defence, instead preferring to pursue more of a passive defence approach.

"The DASA bid that we won in 2019 was all about the invisible battlespace. It was about sparse connectivity, noisy disruption, lots of countermeasures being thrown at you, trying to identify channels, making sure you've got safe connections," he says.

"We've commercialised that, and it's part of our 5G satellite hybrid technology in the built environment. We're applying it, for example, in the rail industry, around rail corridors, and we're working closely with the UK and European Space Agencies."

One defence application example of Massive Analytic’s Artificial Precognition platform is increasing drone manoeuvrability precision.

"With fully autonomous drones, we're using Artificial Precognition to make much more precise manoeuvres effectively. If you imagine a cone where the machine or the AI will put the vehicle in three dimensions, it might have a very wide angle. But what AP does is narrows the angle; it makes it much, much more precise.

"It's a pre-cognitive, cognitive intelligence based on an adaptive AI control circuit that effectively controls machines and with applications in data science and other areas. More recently, we've combined quantum computing and AI to create quantum neural networks capable of calculating in real-time, for example, whether it is safe for a self-driving car to change lanes on a motorway.

"We are working with at least one automotive original equipment manufacturer to test out the concept. The application of quantum AI technologies to defence creates yet another level of military superiority."

If future conflicts involve AI and autonomous technologies, the plus side is that it should keep troops out of harm’s way. However, there are major ethical concerns about the potential of building weapons capable of thinking for themselves.

While leading world governments may sign up to international standards of ethics, there could be severe consequences if terrorists or rogue states launched attacks using AI and autonomous technologies.

"Effectively, you don't need to have soldiers on the battlefield anymore. If you can achieve that, you can go out and kill targets, or you can destroy supply lines and assets with virtually no loss of life to you, and that's the race we're in," adds Frangou.

There are predictions that future wars involving advanced technology will resemble something out of a science-fiction franchise. One such potential threat is swarms of autonomous drones that individually fit on the palm of a hand. Swarms of thousands could make precise targeted attacks; the small size and sheer volume of drones would make them extremely difficult to stop.

But Frangou does not believe we will ever see a 'Terminator' style future where killer robots carry out attacks independent of human control. However, other potential scenarios with AI and automation could have dire consequences for humanity, highlighting the urgent need to develop technologies to counter such threats.

"There are some nightmare, apocalyptic scenarios. Someone – and it will be a person, not a machine – with many geopolitical flashpoints in the world believes that they have achieved such a level of AI superiority – fully deploying and realising that AI superiority – that they can win a nuclear war with a first strike," warns Frangou.

"When deploying that technology, it escalates into tactical nuclear warfare and then escalates into strategic nuclear warfare. I don't mean in the full doomsday scenario, but then you've got the exchange of nuclear weapons.

"You're looking at hundreds of millions of people dead, and then the economics of the world completely changes. And the shift of power changes. Do I believe that's going to happen? I don't know. But we have to win the war. If there is a kind of ‘Cold War’ in the AI space, we have no choice. We must win it."