Countries at Risk for Intrastate Mass Killing 2024–25: Early Warning Project Statistical Risk Assessment Results

IntroductionBack to top

Figure 1. Heat map of estimated risk of new intrastate mass killing, 2024–25

Policy makers face the challenge of simultaneously responding to ongoing mass atrocities and trying to prevent entirely new mass atrocity situations. A critical first step toward prevention is accurate and reliable assessment of countries at risk for future violence. The Early Warning Project’s Statistical Risk Assessment uses publicly available data and statistical modeling to produce a list of countries ranked by their estimated risk of experiencing a new episode, or onset, of intrastate mass killing.  This report aims to help identify countries where preventive actions may be needed. Earlier identification of risk broadens the scope of possible preventive actions.

In essence, our statistical model identifies patterns in historical data to answer the following question: Which countries today look most similar to countries that experienced mass killings in the past, in the year or two before those mass killings began? The historical data include basic country characteristics, as well as data on governance, war and conflict, human rights and civil liberties, and socioeconomic factors. 

This report highlights findings from our Statistical Risk Assessment for 2024–25, focusing on:

• Countries with the highest estimated risk of a new mass killing in 2024 or 2025
• Countries where estimated risk has been consistently high over multiple years
• Countries where estimated risk has increased or decreased significantly from our last assessment
• Countries with unexpected results

We emphasize that this assessment is just one tool. It is meant to be a starting point for discussion and further research, not a definitive conclusion. We aim to help governments, international organizations, and nongovernmental organizations determine where to devote resources for additional analysis, policy attention, and, ultimately, preventive action. We hope that this report and our Early Warning Project as a whole inspire governments and other organizations to invest in their own early warning capabilities.

---

Definition: intrastate mass killingBack to top

By our definition, an intrastate mass killing occurs when the deliberate actions of armed groups in a particular country (including but not limited to state security forces, rebel armies, and other militias) result in the deaths of at least 1,000 noncombatant civilians in that country over a period of one year or less. The civilians must also have been targeted for being part of a specific group. Mass killing is a subset of “mass atrocities,” which we define more generally as “large-scale, systematic violence against civilian populations.” 

---

Understanding these resultsBack to top

Before discussing the results, we underscore five points about interpreting this Statistical Risk Assessment: 

First, as a statistical matter, mass killings are rare. On average, about 1.4 percent of countries see a new mass killing in any given year—that means about two countries. Our risk model predicts a similar number of new episodes of mass killing, so the average two-year risk estimate produced by our model is 1.6 percent. Out of 168 countries, only Chad, Sudan, Burma, and Pakistan have two-year risk estimates greater than 10 percent, or about a one in 10 chance of experiencing a new mass killing in 2024 or 2025.

Second, our model is designed to assess the risk of a new mass killing, not of the continuation or escalation of ongoing episodes. Much of the Simon-Skjodt Center’s work spotlights ongoing mass atrocities and urges lifesaving responses. We focus here on the risk of new mass killing to help fill an analytic gap that is critical to prevention. This feature is especially important to bear in mind when interpreting results for countries that are currently experiencing mass killings, including seven in the top 15 of this assessment (see Figure 2 and our website for a full list of countries). For these countries, our assessment should be understood as an estimate of the risk that a new mass killing event would be launched by a different perpetrator or targeting a different civilian group in 2024 or 2025. While it is important to focus on countries already experiencing mass killing and at high risk of a new onset, it is also essential to focus additional attention on high-risk countries with no ongoing episodes. Regardless of their ranking in this assessment, cases of ongoing atrocities demand urgent action (see Figure 4 and our website for the Early Warning Project’s complete list of ongoing mass killings).

Third, we only forecast intrastate mass killings, where the perpetrator and targeted civilian group reside in the same country (see Box 1). For simplicity, we will refer to it hereafter as "mass killing." This risk assessment does not forecast situations in which an armed group (state or nonstate) residing in one country attacks civilians in another country’s territory. The decision to exclude interstate or cross border mass killings from our model does not involve a value judgment about the moral or practical significance of such atrocities, only a pragmatic judgment about what we are able to forecast more reliably. 

Fourth, readers should keep in mind that our model is not causal: the variables identified as predicting higher or lower risk of mass killings in a country are not necessarily the factors that drive or trigger atrocities. For example, a large population does not directly cause mass atrocities; however, countries with large populations have been more likely to experience mass killing episodes in the past, so this factor helps us identify countries at greater risk going forward. We make no effort to explain these kinds of relationships in the data, some of which can seem perplexing; we only use them for their predictive value. An important consequence of the non-causal nature of these forecasts is that actions aimed at addressing risk factors identified in the model would not necessarily be effective in mitigating the risk of mass atrocities; this assessment does not seek to evaluate atrocity prevention policy prescriptions. For example, although our model finds that countries that have not accepted the First Optional Protocol to the International Covenant on Civil and Political Rights are at greater risk of experiencing mass killings than are other countries, this does not imply that action to encourage ratification of the Optional Protocol would help prevent mass killings. Our risk assessment is meant to be a starting point for discussion and further research, directing policy makers and other practitioners to the countries that merit additional analysis to determine how to help prevent atrocities.

Fifth, this assessment is based on available data reflecting conditions as of the end of 2023. Events that occurred in 2024, such as large-scale violence against protestors in Bangladesh or Bashar al-Assad's removal from power in Syria, are not reflected in country risk estimates. Our assessment relies on publicly available data that are reliably measured for nearly all countries in the world, regularly updated, and historically available going back many years. Most of these data are published a few months after the end of the calendar year to which they refer. As a result, we are able to publish our forecasts for 2024–25, based on data on 2023, only in late 2024.

MethodsBack to top

To produce this assessment, we employ data and statistical methods designed to maximize the accuracy and practical utility of the results. Our model assesses the risk for onset of both state-led and nonstate-led mass killings over a two-year period.

DataBack to top

The data that inform our model come from a variety of sources. On the basis of prior empirical work and theory, we selected more than 30 variables, or risk factors, as input for our statistical model (see the discussion of our modeling approach below). All data used in our model are publicly available, regularly updated, and available without excessive delay. The data also have, in our estimation, minimal risk of retrospective coding bias (coded in a manner influenced by either the presence or absence of observed mass killings), cover all or almost all countries in the world, and go back at least to 1960 (but ideally to 1945). We include variables reflecting countries’ basic characteristics (e.g., geographic region, population); socioeconomic measures (e.g., changes in gross domestic product per capita); measures of governance (e.g., restrictions on political parties); levels of human rights (e.g., freedom of discussion); and records of violent conflict (e.g., battle-related deaths, ongoing mass killings). Alongside the model, we publish a data dictionary, which includes descriptions of all the variables (also referred to as risk factors) included in our model. We also make the model and all data available on our GitHub repository. The only data set the Early Warning Project maintains is that of new and ongoing mass killing events. Although we minimize changes to the data and model in between our every-five-year revision process, we made minor adjustments to improve data completeness this year.

Despite our efforts to include a wide range of relevant variables, data availability and quality are continuing challenges. Some risk factors that might be useful predictors (e.g., dangerous speech) are not included in the model because data meeting the above criteria are unavailable. Additionally, in countries or remote regions where information is limited and/or where governments deliberately restrict access to international observers, such as in Afghanistan, Burma, China’s Xinjiang Uyghur Autonomous Region, Ethiopia, or North Korea, existing data might not fully reflect conditions in the country. 

Modeling approachBack to top

Our modeling approach is described in detail on our website. We use a logistic regression model with “elastic-net” regularization, which is one approach that aids in avoiding “overfitting” the model to the data. Based on a set of more than 30 variables and data on mass killing going back to 1960, the algorithm identifies predictive relationships in the data, resulting in an estimated model. We then apply this model to recent data (from 2023 for the 2024–25 assessment) to generate forecasts. The project includes all internationally recognized independent states with populations of more than 500,000. The model automatically selects which variables among these are useful predictors of a mass killing onset. We emphasize that these risk factors should not be interpreted as causes or “drivers” of risk but simply as correlates of risk that have proven useful in forecasting. Indeed, many of these variables may be useful predictors not because they cause mass killing to be more likely, but because they indirectly serve as proxies for other factors that do.

AccuracyBack to top

We assessed the accuracy of this model in ways that mimicked how we use its results: We built our model on data from a period of years and then tested its accuracy on data for later years. We compared model-based forecasts to actual mass killing onsets (or non-onsets) in the relevant years. This provided an “out-of-sample” test of how well the model performed on data that weren’t already used to train it. Our results indicate that in any given year, about two out of every three countries that later experienced a new mass killing ranked among the top 30 countries.

Highlights from the 2024–25 Statistical Risk Assessment Back to top

Our model generates a single risk estimate for each country, representing the estimated risk for a new state-led or nonstate-led mass killing. Figure 2 displays the estimated risk in 2024 or 2025 for the 30 highest-ranked countries. For every country in the top 30, we recommend that policy makers consider whether they are devoting sufficient attention to addressing the risks of mass atrocities occurring within that country. Strategies and tools to address atrocity risks should, of course, be tailored to each country’s context.

Further qualitative analysis is needed to understand the drivers of risk in a given situation, the mass atrocity scenarios that could be deemed plausible, and the resiliencies that could potentially be bolstered to help prevent future atrocities. This kind of deeper qualitative assessment is exemplified in Early Warning Project reports on Uganda (2024), Indonesia (2022), Côte d’Ivoire (2019), Mali (2018), Bangladesh (2017), and Zimbabwe (2016). Concerned governments and international organizations should consider conducting their own assessments of countries at risk, which should suggest where adjusting plans, budgets, programs, and diplomatic strategies might help prevent mass killings in high-risk countries. For example, the 2022 US Strategy to Anticipate, Prevent, and Respond to Atrocities outlines commitments by the US government to conduct these types of assessments on identified priority countries, guided by the US Atrocity Risk Assessment Framework. Because these qualitative assessments are resource intensive, policy makers should prioritize that type of analysis on countries whose risk estimate is relatively high, according to this Statistical Risk Assessment, and where opportunities for prevention exist.

In the paragraphs below, we discuss each country’s statistical risk ranking and note instances of ongoing violent conflict, group-targeted human rights abuses, and significant events that pose risk for major political instability. These brief summaries include information that goes beyond the data in our statistical model, but they are not intended to provide a comprehensive analysis of factors contributing to atrocity risk. Rather, they are intended to serve as starting points for those who are interested in deeper qualitative analysis. For each country, we also identify the specific factors that account for the risk estimates from our model (see “Methods” for more detail on the risk factors in the model) and note whether the country is experiencing an ongoing mass killing.

Figure 2. Top 30 countries by estimated risk of new intrastate mass killing, 2024–25 

Note: * Indicates an ongoing state-led mass killing; ° indicates an ongoing nonstate-led mass killing. Some countries have multiple ongoing episodes of one or both types (e.g., Burma/Myanmar has three ongoing state-led mass killings; Nigeria has an ongoing state-led and an ongoing nonstate-led mass killing). Risk-based ranking is in parentheses. The probabilities displayed here are associated with the onset of an additional mass killing episode. See the full list of ongoing mass killings on our website.

---

KEY QUESTIONS USERS SHOULD ASK

The results of this risk assessment should be a starting point for discussion and analysis of opportunities for preventive action. For countries in each of the following categories, we recommend asking certain key questions to gain a fuller understanding of the risks and adequacy of policy response and to identify additional useful lines of inquiry.

Highest-risk and consistently high-risk

• Are the risks of large-scale, systematic attacks on civilian populations in the country receiving enough attention?
• What additional analysis would help shed light on the level and nature of atrocity risk in the country?
• What kinds of crises or events (e.g., coups, elections, leadership changes, protests, etc.) might spark large-scale violence by the government or nonstate actors?

Increasing risk

• What events or changes explain the big shifts in estimated risk?
• Have there been additional events or changes, not yet reflected in the data, that are likely to further shift the risk?
• Is the increase part of an ongoing trend?

Unexpected results

• What accounts for the discrepancy between the statistical results and experts’ expectations?
• What additional analysis would help shed light on the level and nature of atrocity risk in the country?

---

Highest-risk countriesBack to top

• Chad (Rank 1): Chad has ranked among the 30 highest-risk countries in our assessment for multiple years, but this is the first time it has been in the top three. The country has faced widespread instability since the April 2021 death of its president of three decades, Idriss Déby, which was followed by a military coup. The junta violently repressed mass protests in October 2022 that were sparked by frustrations at the much-delayed transition back to civilian control. Chadian security forces are estimated to have killed over 100 individuals and arrested nearly 1000 in the brutal crackdown. Ahead of the presidential election in May 2024, Chad experienced widespread political violence and repression of dissent, including the killing of prominent opposition leader Yaya Dillo by security forces. Although Mahamat Idriss Déby Itno—the son of the former president—won the presidential election, many groups have expressed concerns that the election was neither free nor fair. In the wake of the election, the government detained hundreds of people protesting the results, and at least 12 people died in violence surrounding the vote. In addition to political instability, Chad continues to experience spillover effects from the war in Sudan, with more than 600,000 refugees entering the country from Sudan since April 2023. With many Zaghawa and Arab communities spanning the Chadian-Sudanese border, the Chadian government faces heightened tensions as it balances competing demands from groups tied to the conflict in Sudan. Violence between nomadic herders and sedentary farmers in central and southern Chad continues to pose a threat to civilians. At least 1,230 people were killed and 2,200 injured as a result of settler-herder violence from 2021 to mid-2024. According to our model, the factors accounting most for Chad’s high risk estimate are that it experienced a coup in the past five years (2021), that political killings are frequently approved of or incited by top leaders of government, its history of mass killing, and its high infant mortality rate.

Figure 3.

• Sudan (Rank 2): Sudan has ranked among the ten highest-risk countries for multiple years. Violent conflict between the Sudan Armed Forces (SAF), the Rapid Support Forces (RSF), and respective allied ethnic militias has continued to escalate since it began in April 2023. Estimates of civilian fatalities range widely, but according to ACLED, as of October 2024, nearly 7000 civilians have been killed and 8.16 million people displaced in the conflict. Both the SAF and the RSF have been responsible for violence toward civilians through indiscriminate bombings, direct attacks, and the intentional blocking of humanitarian aid. As fighting has spread, local ethnic militias have been drawn into the conflict and civilians are increasingly targeted based on their identity or perceived support for or affiliation with different forces. There continue to be widespread and systematic reprisal attacks and rampant sexual and gender-based violence (largely associated with RSF forces). Civilians in Darfur have endured some of the worst violence, with RSF and allied militia forces intentionally targeting civilians—primarily non-Arab—based on their identity and perceived support to the SAF. The Early Warning Project considers the RSF’s attacks in Darfur to constitute a nonstate-led mass killing that began in 2023; this episode is distinct from the state-led mass killing of non-Arab ethnic groups in Darfur that has been ongoing since 2003. The current risk assessment relates to the likelihood of a new and distinct mass killing, not the continuation or escalation of these ongoing episodes. As the conflict intensifies in other regions of Sudan, the RSF has targeted other groups, including, for example, killing 70 people from the Nuba ethnic group in South Kordofan in February 2024 and more recently killing over 120 people in the eastern town of Al-Jazirah in October 2024. According to our model, the factors accounting most for Sudan’s high risk estimate are that it experienced a coup in the past five years (2021), its history of mass killing, and its large number of battle-related death.
• Burma/Myanmar (Rank 3): The Early Warning Project considers Burma to have three distinct, state-led mass killing episodes ongoing (see here for details). Burma’s high ranking this year indicates that the country continues to exhibit many characteristics common among countries that experience new mass killings, as the current risk assessment relates to the likelihood of the onset of a new and distinct mass killing, not the continuation or escalation of an ongoing episode. This year, renewed campaigns by nonstate ethnic armed groups resulted in territorial gains and significant losses for the military. In response, the military intensified its attacks on civilians (namely the Rohingya, Karen, Kachin, Ta’ang, Mon, Lisu, and Shan). The military has killed at least 5,350 civilians since the 2021 coup. The situation in Rakhine State has deteriorated as both nonstate and state forces target civilians. The Arakan Army (AA), which has seized control over a large part of the Rakhine state, has reportedly killed Rohingya civilians and burned and looted Rohingya villages, forcing many to flee. In one instance, the AA reportedly killed nearly 200 civilians in a drone strike as the latter attempted to flee fighting in Maungdaw. Additionally, both the military and ethnic armed groups have forcibly recruited civilians as soldiers or labor through kidnapping, violence, and threats. According to our model, the factors accounting most for Burma’s high risk estimate are that it experienced a coup in the past five years (2021), that political killings are frequently approved of or incited by top leaders of government, and its history of mass killing.

Countries in the top ten that are not discussed in this year’s report are India, Guinea, and Yemen. To learn more about the factors that contributed to the high risk estimate of any of these countries, visit the Early Warning Project’s homepage and use the “select a country” dropdown in the top right corner.

---

BURMA: THE DIFFERENCE BETWEEN NEW ONSETS AND CONTINUING INTRASTATE MASS KILLING

Readers are sometimes unsure how to interpret the risk assessment results for a country, such as Burma, that is already experiencing a mass killing episode.

The percentage risk and ranking for each country represents the estimated probability that a new onset of mass killing begins in that country—that either a new perpetrator group emerges and kills more than 1,000 civilians of a specific group, or an existing perpetrator group begins targeting a new group of civilians—not that an existing mass killing continues. This decision follows the project’s goal to provide early warning before large-scale killings begin, while opportunities for prevention are greatest.

In Burma, there were three ongoing mass killings as of the end of 2023: a state-led mass killing against civilians suspected of opposing the military junta since 2021, a state-led mass killing against Rohingya civilians since 2016, and a state-led mass killing against ethnic minority groups in the country’s east since 1948. Burma’s risk and ranking (10.1 percent risk and 3rd rank) are estimates of the likelihood that a different civilian group would be targeted and/or a new perpetrator group would emerge in 2024 or 2025, not the continuation or escalation of an ongoing episode. Burma’s high ranking this year indicates that the country continues to exhibit many characteristics common among countries that experience new mass killings. While our assessment is not able to predict what a new onset of mass killing in Burma might look like, additional analysis can help shed light on plausible scenarios. Given the state's already broad targeting of civilian groups, it may be unlikely (but not impossible) that the government would expand its focus to new groups. However, with numerous nonstate armed groups active around the country, there remains a possibility that Burma might experience a nonstate-led mass killing event.

See the Museum’s website for more information about the crisis in Burma, efforts to bring it to an end, and to promote justice and accountability.

---

Consistently high-risk countriesBack to top

In addition to Chad, Sudan, and Burma, a few other countries have appeared near the top of our rankings for several years. We highlight the following three:

• Pakistan (Rank 4): Pakistan has ranked among the ten highest-risk countries in our assessment since 2014 and the five highest-risk countries since 2018. Pakistan faces a number of security challenges, including attacks by the Tehrik-e-Taliban Pakistan (TTP), a nonstate armed group responsible for a mass killing episode that the Early Warning Project has judged as ongoing since 2001. The current risk assessment relates to the likelihood of the onset of a new and distinct mass killing, not the continuation or escalation of this episode. Violence by nonstate armed groups has sharply escalated over the past few years; August 2024 was the “deadliest month in six years.” This year, attacks by the TTP, the increasingly lethal Baloch Liberation Army (BLA), and other nonstate groups killed nearly 500 civilians. The political landscape remains tumultuous. The legitimacy of the February 2024 national elections has been called into question due to numerous accusations of electoral interference, political repression, and “irregularities” in the voting process. In October 2024, Parliament passed a controversial amendment increasing government influence over judicial appointments, including the chief justice. Religious minorities—including Ahmadi Muslims, Christians, Hindus, Shia and Sikhs—are subject to both official and societal discrimination and are occasionally targeted in religiously motivated attacks. Afghan refugees—including the more than half a million who sought shelter in Pakistan since the Taliban’s August 2021 return to power in Afghanistan—have emerged as a particularly vulnerable group. Over the past year, the government announced—and then subsequently amended—a plan to deport millions of Afghans residing in the country; Pakistani security forces have been accused of committing grave abuses against Afghans in order to coerce them into returning to Afghanistan, where many are at high risk of persecution by the de facto authorities. Separately, in Kurram, clashes between Sunni and Shia tribes over land disputes have led to numerous deaths, including 82 people killed over three days in November 2024. According to our model, the factors accounting most for Pakistan’s high risk estimate are its large population, history of mass killing, and large number of battle-related deaths.
• Ethiopia (Rank 7): Ethiopia has consistently ranked among the top ten highest-risk countries in our assessment. State-led mass killings against Tigrayan civilians and against Oromo civilians were both ongoing as of the end of 2023. The current risk assessment relates to the likelihood of a new mass killing, not the continuation or escalation of an ongoing one. Ethiopia’s high ranking indicates that the country continues to exhibit many characteristics common among countries that experience new mass killings. Although the Ethiopian government and the Tigray People’s Liberation Front signed a Cessation of Hostilities Agreement in November 2022, risks still remain in the Tigray region, in addition to ongoing violent conflict between government forces and militias in the Amhara and Oromia regions. The United Nations reported a sharp increase in violence in 2023 from 2022—at least 1,351 civilians were killed in 2023 in attacks carried out by government forces, Eritrean troops, anti-government militias, and other actors. Both state and nonstate actors appear to be intentionally targeting civilians. For example, government forces killed at least 89 civilians in Merawi in January 2024 in a particularly deadly episode. Increasing tensions with neighboring countries and spillover effects from the war in Sudan pose additional risks to Ethiopia’s political stability. According to our model, the factors accounting most for Ethiopia’s high risk estimate are its history of mass killing, large population, and large number of battle-related deaths.
• Afghanistan (Rank 8): Afghanistan has ranked among the ten highest-risk countries in our assessment since 2015. Following the Taliban’s takeover in August 2021, the de facto authorities have significantly eroded rights and freedoms, including by severely restricting women’s rights and suppressing political opposition. Despite an overall reduction in conflict-related violence since the takeover, over 1,000 civilians were killed between August 15, 2021, and May 2023—most of which can be attributed to the nonstate armed group the Islamic State of Khorasan Province (ISIS-K). Afghanistan’s Hazara community has faced targeted violence, with ISIS-K claiming responsibility for at least 17 attacks that have wounded or killed more than 700 people since August 2021. The Taliban continues to restrict women’s rights, severely curtailing their freedom to work, travel, pursue education beyond the sixth grade, and, under a new law, even speak in public. Ethnic and religious minorities face increasing marginalization and discrimination, and freedom of the press and other media outlets is all but non-existent. Additionally, despite the Taliban’s promise not to punish those affiliated with the former Afghan government, the UN documented 14 extrajudicial killings and more than 100 human rights violations against this group within the first nine months of 2024. According to our model, the factors accounting most for Afghanistan’s high risk estimate are that political killings are frequently approved of or incited by top leaders of the government and its history of mass killing.

---

WHAT ABOUT CROSS-BORDER OR INTERSTATE MASS KILLING?

The Early Warning Project’s definition of intrastate mass killing excludes situations in which an armed group (state or nonstate) residing in one country attacks civilians in another country’s territory. The only exceptions to this rule are situations where we can document substantial and close coordination in killing operations between the foreign armed group and the government of the state where the targeted civilian group resides. 

This means our definition of mass killing does not include Russian forces’ deliberate targeting of civilians in Ukraine, civilian killings in Yemen perpetrated by the Saudi-led coalition, or civilian killings in the war between Israel and Hamas. 

The decision to exclude these mass killings does not involve a value judgment about the moral or practical significance of atrocities perpetrated during wars between states, international terrorist actions, and other international military operations, only a pragmatic judgment about what we are able to forecast more reliably. 

---

Significant shifts in rankingBack to top

We highlight two countries that moved up in our rankings substantially between the 2023–24 and 2024–25 assessments.

• Burkina Faso (Rank 6): Burkina Faso rose from 29th last year to 6th this year, primarily because this is the first year we recorded that the country had experienced a mass killing onset (in 2019) and because, according to V-Dem, it now bans some political parties. This year, the Early Warning Project determined that there has been an ongoing nonstate-led mass killing of civilians by militant nonstate Islamist groups since 2019. The current risk assessment relates to the possibility of a new and distinct nonstate-led or state-led episode beginning, not to the ongoing episode continuing or escalating. In 2024, the conflict between state forces and Islamist nonstate armed groups, such as the Jama’at Nusrat al-Islam wal-Muslimin (JNIM) and the Islamic State in the Greater Sahara (ISGS), continued. Notably, the JNIM claimed responsibility for an attack on civilians in Barsalogho in August that killed 200 people. Islamist groups have also blocked food and humanitarian aid to civilians, leading to starvation. Additionally, state-led forces have targeted civilians in their military operations. On February 25, the Burkina Faso military executed at least 223 civilians, including 56 children, in the villages of Nondin and Soro for their suspected collaboration with Islamist groups. In May, the military killed over 270 civilians in Yagha and Komondjari provinces. Along with extrajudicial killings, Burkina Faso security forces have reportedly used intimidation, arbitrary detention, forced disappearances, and unlawful recruitment to suppress journalists, human rights defenders, political opponents, and dissenters. The Burkina Faso military has also allegedly retaliated against prosecutors and judges by unlawfully conscripting them into service. According to our model, the factors accounting most for Burkina Faso’s high risk estimate are the recent coup d’état (2022), its history of mass killing, and high battle-related deaths.
• South Sudan (Rank 18): South Sudan moved up considerably in our risk assessment, from 86th in 2023–24 to 18th in 2024–25. The shift is surprisingly large given that the data on South Sudan’s risk factors has changed very little since last year. As sometimes happens when forecasting a rare outcome using a large number of variables, minor changes to the model account for this significant shift. This volatility, together with concerning recent developments, suggests that additional analysis of South Sudan is merited. Despite the 2019 ceasefire, South Sudan continues to experience ongoing armed conflict, human rights violations, and a severe humanitarian crisis. According to the UN Mission for South Sudan, violent conflict between nonstate armed groups appears to be on the rise and “was the leading cause of harm affecting civilians” in 2023. The Early Warning Project already considers there to be two mass killing episodes—one state-led and one nonstate-led—ongoing in South Sudan since 2013. The current risk assessment relates to the possibility of a new and distinct nonstate-led or state-led episode beginning, not to the ongoing episodes continuing or escalating. In September 2024, the government of South Sudan announced a two-year extension of the transitional period, postponing the slated December 2024 national elections until December 2026. Reactions to the extension have been mixed, with some actors supporting it as an opportunity to mitigate risks before elections and others expressing frustration or disappointment with the government for delaying a democratic transition. The ongoing war in neighboring Sudan continues to contribute to tensions, as the influx of Sudanese refugees into South Sudan strains resources and exacerbates existing local disputes. A detailed qualitative assessment is necessary to help understand the nature and severity of atrocity risks, whether they be from the escalation of an ongoing episode or the start of a new episode. According to our model, the factors accounting most for South Sudan’s high risk estimate are its history of mass killing, high infant mortality rate, and that it bans some political parties.

---

EXPLORING CHANGES TO A COUNTRY’S RISK FACTORS: THE EXAMPLE OF BANGLADESH

The data used to produce this assessment are from 2023 (published by most sources in early-to mid-2024). This means that changes that occurred in 2024 are not captured in this risk assessment. To enable users to examine how such changes might affect a country’s risk estimate and ranking, our online platform has an interactive data tool. The tool allows users to explore how changes to a country’s risk factors would affect its risk of mass killing, holding all other variables constant. Users may want to:

1. See what a country’s risk and ranking would be if we were to observe some different set of values of its risk factors (e.g., though no war broke out and battle-related deaths were zero, what if we instead saw a large number of battle-related deaths?)
2. Manually update country risk based on known changes (e.g., knowing that a coup occurred in a country, users can see how a change in that variable would affect the risk and ranking)
3. Adjust risk factor values where users disagree with a data source’s coding judgments

For example, in 2024–25, Bangladesh ranks 13th, with a 4.9 percent estimated risk, or a one in 20 chance of a new mass killing. This assessment is based on 2023 data. However, someone following Bangladesh may suspect that events over the course of 2024—namely, the mass anti-government protests and the subsequent violent crackdowns by state forces that killed hundreds of civilians—may impact that risk.

Using the tool, we see, for example, that if political killings were to become systematic and incited or approved by top government leaders, Bangladesh would rank 6th and its new risk estimate would go up to 8 percent, or about a one in 13 chance of a new mass killing.

---

Unexpected resultsBack to top

Global statistical risk assessments can help by identifying countries whose relatively high (or low) risk estimates surprise regional experts. In cases where our statistical results differ substantially from expectations, we recommend conducting deeper analysis and revisiting assumptions. The purpose of this analysis is not to pit qualitative analysts and statistical models against one another but rather to deepen our understanding of risk in the country in question. We highlight two countries that, in our informal judgment, fall into this category. 

• Venezuela (Rank 81): Venezuela’s ranking of 81st may strike some observers as surprisingly low, considering the country’s ongoing political and human rights crisis. Nearly eight million Venezuelans have fled the country since 2014, when the government violently cracked down on protests. A 2020 UN Fact Finding Commission concluded there were reasonable grounds to believe that the government had committed crimes against humanity, including murder, torture, rape, and enforced disappearances. Most recently, the highly contested presidential election in July 2024 did not meet international democratic standards, according to a group of UN experts and the Carter Center. Following the elections, mass protests erupted and were met with violence, largely by state forces, including organized intimidation of opposition supporters, the mass arrest of more than 2,200 political dissidents, and the killing of at least 23 protestors or bystanders. Venezuela’s human rights crisis has included widespread violations but has so far not included large-scale killing. According to our model, the factor accounting most for Venezuela’s lower risk estimate is its geographic region (Americas).
• Haiti (Rank 96): Haiti fell further in rank from 59th in 2023–24 to 96th in 2024–25. To observers tracking the humanitarian crisis and high levels of gang violence in Haiti, this ranking may seem unexpectedly low. Although a truce between nonstate armed groups decreased clashes between these groups, as well as the number of attacks on civilians in 2024, the number of civilian fatalities remained largely unchanged from 2023. Civilians have been the target of intentional gang-related violence, including during an attack that killed at least 115 people in October 2024. There are also reports of intentional and escalating use of sexual violence as a means of intimidation and control by gangs, including against children. The situation remains volatile, as seen by the gang attacks on government facilities, including the March 2024 jailbreak of over 3,000 inmates from two prisons, which forced the resignation of Prime Minister Ariel Henry in April 2024. Though the scale of violence in Haiti is consistently high, armed groups have not deliberately and systematically targeted a specific group of civilians, and those responsible include multiple armed groups (gangs and state security forces) operating independently. According to our model, the variables that account for Haiti’s low ranking are the lack of variation in respect for civil liberties across different parts of the country (according to V-Dem) and its geographic region (Americas).

ConclusionBack to top

Early warning is a crucial element of effective atrocity prevention. The purpose of our statistical risk assessment is to provide one practical tool to the public for assessing risk in countries worldwide. This tool should enable policy makers, civil society, and other analysts to focus attention and resources on countries at highest risk, especially those not currently receiving sufficient attention. 

This quantitative assessment is designed to serve as a starting point for additional analysis. Governments and international organizations have developed and implemented tools for qualitative atrocity risk assessments. We see the application of such tools as complementary to our statistical analysis. In-depth assessments that consider both qualitative and quantitative analyses should in turn spur necessary adjustments in strategic plans, budgets, programs, and diplomatic strategies toward high-risk countries. By combining these approaches—global risk assessment, in-depth country analysis, and preventive policy planning—we have the best chance of preventing future mass atrocities. 

Figure 4.