Using AI to anticipate future cyber attacks
Cyber attacks are becoming more frequent and sophisticated, yet most cyber security systems can only respond once an attack is already underway. This research led by Professor Paul Yoo, uses artificial intelligence (AI) to analyse large amounts of information, including security reports, system logs and scientific publications, to identify emerging threats and predict how cyber attacks may develop in the future.
While AI has improved the speed and accuracy of detecting ongoing attacks, far less progress has been made in forecasting threats years in advance. Funded through NATO's Science for Peace and Security (SPS) Programme, this project aims to address that gap by developing AI methods that can predict long-term trends in cyber attacks and defensive technologies, giving organisations and governments more time to prepare before threats emerge.
Why is this research important?
Predicting cyber threats years in advance remains one of the biggest challenges in cyber security. While AI has significantly improved the speed and accuracy of detecting attacks that are already underway, forecasting future threats remains much more difficult.
This research aims to change that by helping organisations move from reacting to cyber attacks to preparing for them. Earlier warnings would give governments, businesses and critical services more time to strengthen their defences, invest in the right technologies and work together to reduce risks.
Today, long-term cyber security forecasting relies heavily on a small number of highly experienced experts. By developing AI tools that can support this expertise, the project could make strategic cyber security planning more widely available and effective.
How is the research being conducted?
The project brings together researchers and cyber security specialists from the UK and Australia. Working with experts at Royal Melbourne Institute of Technology (RMIT University), the team combines knowledge of artificial intelligence with practical experience of cyber security operations.
The researchers also work closely with government and industry partners, including specialists from Samsung and the UK Cabinet Office. These collaborations help ensure that the research addresses real-world challenges and produces tools that organisations can use in practice.
What do you hope the impact of the research will be?
The research aims to help governments, businesses and other organisations anticipate how cyber threats and defensive technologies may evolve over the coming years.
For governments, this could support national cyber security planning and help prioritise investment in future capabilities. For businesses, it could provide evidence to guide spending on new technologies and cyber defences before emerging threats become widespread.
“Ultimately, the goal is to create a more resilient digital environment by helping organisations anticipate and prepare for cyber threats before they occur.”
- Full project title: Explainable Forecasting of Cyber Threats and Pertinent Alleviation Technologies
- Project funding: £155,000
- Funder: North Atlantic Treaty Organization (NATO)
- Dates of research: 1 April 2025 to 31 March 2027
- Lead researcher: Professor Paul Yoo