This paper recently announced that a 15-kilometer crater, hidden beneath a layer of volcanic rock in southern Laos, may be the source of the Australasian strewn field. But more research will be required before we know for certain.

This is among the most comprehensive books on the topic of civilization-ending threats. While my video is explicitly about human extinction, this book also covers catastrophic near-extinction events. However, due to its age, some of the information is a bit outdated. It's a good read nonetheless.

The claim that an impact with a 10-kilometer-sized (or larger) object would be more energetic than the simultaneous detonation of billions of atomic bombs refers to the two bombs dropped on Hiroshima and Nagasaki during World War II. This document states that Little Boy (Hiroshima) had a yield of 6.276×10¹³ J while Fat Man (Nagasaki) had a yield of 8.786×10¹³ J. Meanwhile, reference #11 estimates the kinetic energy of the Chicxulub impact was between 1.3×10²⁴ J and 5.8×10²⁵ J. That means the impact was between 15-924 billion times more energetic than the bombs dropped on Hiroshima and Nagasaki.

According to this paper, the estimated size of the Chicxulub impactor ranges between 10-80 kilometers depending on factors such as velocity, angle of impact, and composition of the impactor. However, most other estimates land somewhere near the lower end of that range.

The Small-Body Database operated by NASA's Jet Propulsion Laboratory contains information about most asteroids and comets. This reference will query the database for all near-earth objects. When this video was made, the query returned 25 016 results, but new objects are discovered every year. Only a small fraction of these objects have Earth-crossing orbits. The vast majority do not pose a threat.

The Small-Body Database operated by NASA's Jet Propulsion Laboratory contains information about most asteroids and comets. This reference will query the database for damocloids as defined by David Jewitt in reference #18. When this video was made, the query returned 184 results.

The percentage of species that went extinct during the five major mass extinctions are difficult to estimate. They can vary quite significantly between sources. The percentages used in the video are based on this paper as it appears to be the most recent.

This book was a huge influence on the making of this video. The author pays special attention to existential risks and not just catastrophic ones. Highly recommended.

The estimated death toll for the Black Death differs a great deal between sources. Furthermore, estimates often differ in scope. Some include all plague outbreaks throughout human history. Others focus exclusively on the Black Death pandemic that primarily ravaged Europe between roughly 1346 and 1353 CE. The "25-60%" and "20-70 million" ranges used in the video were informed by the three sources listed below. Especially reference #51c.

The death toll of the Columbian exchange is virtually impossible to estimate. No one knows how large the indigenous population was before the European colonization of the Americas. The "~55 million" figure used in the video is merely a best guess and does not differentiate between causes of death. But the majority of deaths were likely due to disease.

Much like the Black Death, the death toll for the Spanish Flu is difficult to estimate. Accurate record keeping was not actively practices at the time of the outbreak and was further impeded by World War I. There are also discrepancies between the estimated number of infections and case fatality rates. Estimates for case fatality rates ranges from 2% at the lower end and 20% at the upper end. The figures used in the video were all informed by the sources listed below.

There have been numerous laboratory escapes over the past few decades. The references below are only a few examples. See reference #50 and reference #66 for a more extensive list.

This book is essential reading if you want to gain a better understanding of the potential risks posed by artificial intelligence.