Tipping points to detect social crises

Detection of acceleration in radicalisation and polarisation, which, as was established in Chapter 2.3, could be exacerbated by Earth system destabilisation, can be pursued using similar machine learning and social network analysis approaches applied to user-generated online content (Gaikwad et al., 2022). Conflict early warning systems (CEWS) are well established and researched (Rød et al., 2023). A notable example is the ACLED (Armed Conflict Location & Event Data Project) CAST platform (Conflict Alert System), which is meant to predict violent events up to six months in advance. These CEWS could be enhanced with new ML/AI-based models that can capture coupled climate-conflict-tipping processes (Guo et al., 2023; Guo et al., 2018). 

Finally, ML/AI-based tools are also emerging to develop early warning systems to predict financial crises (Samitas et al., 2020), which, as was established in chapter 2.3, could be triggered by Earth system destabilisation. Near real-time monitoring is also feasible with these types of data and methods, as demonstrated by the GDELT project, which monitors the world’s broadcast, print and web news from around the world in 100 languages for significant events and trends. With respect to ethical questions around surveillance and privacy concerns, it is important to emphasise that early warning systems focus on broad patterns and do not track individuals, so personally identifiable information is not included in these systems.

Bezos Earth Fund University of Exeter logo
Earth Commission Systems Change Lab logo Systemiq logo
Global Tipping Points logo
Share this content