A Framework for Understanding the Weaponisation of the Internet

Over the past few decades, the Internet has increasingly become an important source of news, background, history and fact. It now forms the primary source of information for decision-making input for a growing segment of the population. Furthermore it may also play an important role in forming a world view, influencing political beliefs and fermenting violent behaviour in some. Credible evidence exists that individuals, state actors and non-state actors, are creating, modifying or promoting information available on the Internet to advance their own agendas. This includes mis-information, dis-information and bias caused by valid information used in the wrong way or at the wrong time, and results in the weaponisation of the Internet. Governments and regulators urgently require tools to detect when the internet is being weaponised, assess its effect on citizens, and ultimately to either prevent or to counter the threat that this causes. This paper proposes a novel framework to model the weaponisation of the internet from a population statistic sense – the level of understanding that regulators and governments will operate at. The model considers threat actors who cause threats that are propagated to targets within a susceptible population over Internet channels. It includes models for the susceptibility and the response of population entities to threats, and the particular cases of amplified or passed-on threats. Echo chambers, the result of an amplification loop, are considered, as are the use of generative AI tools such as modified large language models as both threat originators and threat recipients.