Each month, The Existential Risk Research Assessment (TERRA) uses a unique machine-learning model to predict those publications most relevant to existential risk or global catastrophic risk. The following are a selection of those papers identified this month.
Please note that we provide these citations and abstracts as a service to aid other researchers in paper discovery and that inclusion does not represent any kind of endorsement of this research by the Centre for the Study of Existential Risk or our researchers.
We look at classifying extinction risks in three different ways, which affect how we can intervene to reduce risk. First, how does it start causing damage? Second, how does it reach the scale of a global catastrophe? Third, how does it reach everyone? In all of these three phases there is a defence layer that blocks most risks: First, we can prevent catastrophes from occurring. Second, we can respond to catastrophes before they reach a global scale. Third, humanity is resilient against extinction even in the face of global catastrophes. The largest probability of extinction is posed when all of these defences are weak, that is, by risks we are unlikely to prevent, unlikely to successfully respond to, and unlikely to be resilient against. We find that it’s usually best to invest significantly into strengthening all three defence layers. We also suggest ways to do so tailored to the classes of risk we identify. Lastly, we discuss the importance of underlying risk factors – events or structural conditions that may weaken the defence layers even without posing a risk of immediate extinction themselves.
The study of existential risk — the risk of human extinction or the collapse of human civilization — has only recently emerged as an integrated field of research, and yet an overwhelming volume of relevant research has already been published. To provide an evidence base for policy and risk analysis, this research should be systematically reviewed. In a systematic review, one of many time-consuming tasks is to read the titles and abstracts of research publications, to see if they meet the inclusion criteria. We show how this task can be shared between multiple people (using crowdsourcing) and partially automated (using machine learning), as methods of handling an overwhelming volume of research. We used these methods to create The Existential Risk Research Assessment (TERRA), which is a living bibliography of relevant publications that gets updated each month (www.x-risk.net). We present the results from the first ten months of TERRA, in which 10,001 abstracts were screened by 51 participants. Several challenges need to be met before these methods can be used in systematic reviews. However, we suggest that collaborative and cumulative methods such as these will need to be used in systematic reviews as the volume of research increases.
A major approach to the ethics of artificial intelligence (AI) is to use social choice, in which the AI is designed to act according to the aggregate views of society. This is found in the AI ethics of “coherent extrapolated volition” and “bottom–up ethics”. This paper shows that the normative basis of AI social choice ethics is weak due to the fact that there is no one single aggregate ethical view of society. Instead, the design of social choice AI faces three sets of decisions: standing, concerning whose ethics views are included; measurement, concerning how their views are identified; and aggregation, concerning how individual views are combined to a single view that will guide AI behavior. These decisions must be made up front in the initial AI design—designers cannot “let the AI figure it out”. Each set of decisions poses difficult ethical dilemmas with major consequences for AI behavior, with some decision options yielding pathological or even catastrophic results. Furthermore, non-social choice ethics face similar issues, such as whether to count future generations or the AI itself. These issues can be more important than the question of whether or not to use social choice ethics. Attention should focus on these issues, not on social choice.
We demonstrate that the global cooling resulting from a range of nuclear conflict scenarios would temporarily increase the pH in the surface ocean by up to 0.06 units over a 5-year period, briefly alleviating the decline in pH associated with ocean acidification. Conversely, the global cooling dissolves atmospheric carbon into the upper ocean, driving a 0.1 to 0.3 unit decrease in the aragonite saturation state (Ωarag) that persists for ~ 10 years. The peak anomaly in pH occurs 2 years post conflict, while the Ωarag anomaly peaks 4- to 5-years post conflict. The decrease in Ωarag would exacerbate a primary threat of ocean acidification: the inability of marine calcifying organisms to maintain their shells/skeletons in a corrosive environment. Our results are based on sensitivity simulations conducted with a state-of-the-art Earth system model integrated under various black carbon (soot) external forcings. Our findings suggest that regional nuclear conflict may have ramifications for global ocean acidification.
The idea of superintelligence is a source of mainly philosophical and ethical considerations. Those considerations are rooted in the idea that an entity which is more intelligent than humans, may evolve in some point in the future. For obvious reasons, the superintelligence is considered as a kind of existential threat for humanity. In this essay, we discuss two ideas. One of them is the putative nature of future superintelligence which does not necessary need to be harmful for humanity. Our key idea states that the superintelligence does not need to assess its own survival as the highest value. As a kind of intelligence that is not biological, it is not clear what kind of attitude the superintelligent entity may evolve towards living organisms. Our second idea refers to the possible revelation of superintelligence. We assume that the self-revelation of such entity cannot be random. The metaphor of God as a superintelligence is introduced here as a helpful conceptual tool.
Anthropogenic global warming is one of the most significant existential threats facing the human species. Nonetheless, most individuals largely conduct their lives in a manner that does not fully acknowledge, let alone effectively deal with this threat. This field note argues that both a psychosocial and political-economic approach could offer more in-depth perspectives to understand anthropogenic global warming and potential avenues to investigate it moving forward. In so doing, it is argued that climate change policy recommendations, and associated political action, could benefit from taking into account the dimension of our psyches on an individual and collective level, as well as the political-economic context of anthropogenic global warming.
The 4.2 ka event that occurred during the period from 4 500–3 900 a BP was characterized by cold and dry climates and resulted in the collapse of civilizations around the world. The cause of this climatic event, however, has been under debate. We collected four corals (Porites lutea) from Yongxing Island, Xisha Islands, South China Sea, dated them with the U-series method, and measured the annual coral growth rates using X-ray technology. The dating results showed that the coral growth ages were from 4 500–3 900 a BP, which coincide well with the period of the 4.2 ka event. We then reconstructed annual sea surface temperature anomaly (SSTA) variations based on the coral growth rates. The growth rate-based SSTA results showed that the interdecadal SSTA from 4 500–3 900 a BP was lower than that during modern times (1961–2008 AD). A spectral analysis showed that the SSTA variations from 4 500–3 900 a BP were under the influence of El Niño-Southern Oscillation (ENSO) activities. From 4 500–4 100 a BP, the climate exhibited La Niña-like conditions with weak ENSO intensity and relatively stable and lower SSTA amplitudes. From 4 100–3 900 a BP, the climate underwent a complicated period of ENSO variability and showed alternating El Niño- or La Niña-like conditions at interdecadal time scales and large SSTA amplitudes. We speculate that during the early and middle stages of the 4.2 ka event, the cold climate caused by weak ENSO activities largely weakened social productivity. Then, during the end stages of the 4.2 ka event, the repeated fluctuations in the ENSO intensity caused frequent extreme weather events, resulting in the collapse of civilizations worldwide. Thus, the new evidence obtained from our coral records suggests that the 4.2 ka event as well as the related collapse of civilizations were very likely driven by ENSO variability.