Can artificial intelligence predict the Middle East’s next conflict?

In January 2014, most Iraq analysts knew security in the country was rapidly deteriorating. Beyond general warnings, however, few predicted exactly how bad the situation would become. Within six months, a fanatical terrorist group in the form of ISIS would take over a third of the country, embarking on a brutal campaign of violence. By August, the UN declared Iraq a "level 3 emergency" (the worst kind) as 3 million people fled their homes.

But what if all of this could have been predicted far in advance? Not a general warning that things were getting worse, but rather a detailed prediction outlining the severity of potential conflict and the likely timeline.

Forecasting temporal and spatial aspects of conflict is now the task of researchers at the Turing Institute, the UK’s national institute for data science. It is a problem that has vexed political scientists for decades and to date, most efforts have only succeeded in very broad warnings, ranking states at risk of violent episodes occurring within around one year.

Unsurprisingly, this is because predicting conflict is immensely complex. There is an endless number of inputs for any model. Some analysts focus on environmental factors like drought and food security. Others look at political economy, the interplay of governance and living standards, to provide structural risk analysis.

Complicating this, the sum of data used must be reliable. But in many conflict-affected areas accurate data is hard to come by.

To wade through this maze of information, the Turing Institute has been harnessing artificial intelligence to give policymakers highly specific warnings. This is known as the Global Urban Analytics for Resilient Defence – or GUARD – project.

I asked Weisi Guo, the lead analyst on GUARD, why it is different from previous attempts to foresee conflict. According to Dr Guo, the strength of his institute’s approach involves boosting human analysis with computing power.

“Reasoning in conflict prediction is largely qualitative – involving arguments and debates – or quantitative – involving data and statistical regression (modelling with several variables). We boldly attempt to write logical equations that tie different factors together, and then apply AI to make data-driven predictions.” Behind Dr Guo’s vision is a realisation he made in 2015, when looking at a map of the ancient Silk Road. He was struck by how many of today's most severely conflict areas correspond with the historical web of trade routes that once connected East and West, passing through the Middle East. He then created an algorithm splicing publicly available databases on violent incidents with overland routes, placing physical geography at the heart of his research.

Dr Guo is keen to stress that land routes are not the only component of GUARD. But concentrating on historic “choke-points,” in the flow of goods and people has led his team to some remarkably accurate predictions. In 2017, they were able to predict 76 per cent of the cities in which terror attacks occurred.

Indeed, it is those “junction cities,” where most violence takes place. Mosul, for example, has historically held this status – the city’s name loosely translates as “junction” in English. The geography-violence nexus, Dr Guo emphasises, is particularly relevant in the Middle East. “The Middle East is of course interesting and complex, more so than most places, because it sits at the heart of Eurasia, the super-continent of information, cultural, resources, and people flow, It is by far the most well-connected continent. Being in the middle means [the Middle East] receives a lot more attention, and we use graph theory along with AI to understand which parts of the region are more vulnerable under dynamic political and trade conditions.”

For millennia, the flat expanse of terrain in the vast Tigris and Euphrates river valleys that cross Iraq and Syria has been a blessing and a curse for its inhabitants. Flat terrain enabled rapid movement of goods, accelerating the development of some of the world’s earliest city settlements. These routes were coveted and contested, allowing rival groups to quickly move cavalry into enemy territory.

The writer is a security and political risk analyst focused on Iraq

 

Can artificial intelligence predict the Middle Easts next conflict?

In January 2014, most Iraq analysts knew security in the country was rapidly deteriorating. Beyond general warnings, however, few predicted exactly how bad the situation would become. Within six months, a fanatical terrorist group in the form of ISIS would take over a third of the country, embarking on a brutal campaign of violence. By August, the UN declared Iraq a level 3 emergency (the worst kind) as 3 million people fled their homes. But what if all of this could have been predicted far in advance? Not a general warning that things were getting worse, but rather a detailed prediction outlining the severity of potential conflict and the likely timeline. Forecasting temporal and spatial aspects of conflict is now the task of researchers at the Turing Institute, the UKs national institute for data science. It is a problem that has vexed political scientists for decades and to date, most efforts have only succeeded in very broad warnings, ranking states at risk of violent episodes occurring within around one year. Unsurprisingly, this is because predicting conflict is immensely complex. There is an endless number of inputs for any model. Some analysts focus on environmental factors like drought and food security. Others look at political economy, the interplay of governance and living standards, to provide structural risk analysis. Complicating this, the sum of data used must be reliable. But in many conflict-affected areas accurate data is hard to come by. To wade through this maze of information, the Turing Institute has been harnessing artificial intelligence to give policymakers highly specific warnings. This is known as the Global Urban Analytics for Resilient Defence or GUARD project. I asked Weisi Guo, the lead analyst on GUARD, why it is different from previous attempts to foresee conflict. According to Dr Guo, the strength of his institutes approach involves boosting human analysis with computing power. Reasoning in conflict prediction is largely qualitative involving arguments and debates or quantitative involving data and statistical regression (modelling with several variables). We boldly attempt to write logical equations that tie different factors together, and then apply AI to make data-driven predictions. Behind Dr Guos vision is a realisation he made in 2015, when looking at a map of the ancient Silk Road. He was struck by how many of todays most severely conflict areas correspond with the historical web of trade routes that once connected East and West, passing through the Middle East. He then created an algorithm splicing publicly available databases on violent incidents with overland routes, placing physical geography at the heart of his research. Dr Guo is keen to stress that land routes are not the only component of GUARD. But concentrating on historic choke-points, in the flow of goods and people has led his team to some remarkably accurate predictions. In 2017, they were able to predict 76 per cent of the cities in which terror attacks occurred. Indeed, it is those junction cities, where most violence takes place. Mosul, for example, has historically held this status the citys name loosely translates as junction in English. The geography-violence nexus, Dr Guo emphasises, is particularly relevant in the Middle East. The Middle East is of course interesting and complex, more so than most places, because it sits at the heart of Eurasia, the super-continent of information, cultural, resources, and people flow, It is by far the most well-connected continent. Being in the middle means [the Middle East] receives a lot more attention, and we use graph theory along with AI to understand which parts of the region are more vulnerable under dynamic political and trade conditions. For millennia, the flat expanse of terrain in the vast Tigris and Euphrates river valleys that cross Iraq and Syria has been a blessing and a curse for its inhabitants. Flat terrain enabled rapid movement of goods, accelerating the development of some of the worlds earliest city settlements. These routes were coveted and contested, allowing rival groups to quickly move cavalry into enemy territory. The writer is a security and political risk analyst focused on Iraq