Anthropic shadow: observation selection effects and human extinction risks

Assessing natural global catastrophic risks

The risk of global catastrophe from natural sources may be significantly larger than previous analyses have found. In the study of global catastrophic risk (GCR), one line of thinking posits that deep human history renders natural GCRs insignificant. Essentially, the fact that natural hazards did not cause human extinction at any previous time makes it unlikely that they would do so now. This paper finds flaws in this argument, refines the theory of natural GCR, analyzes the space of natural GCRs, and presents implications for decision-making and research. The paper analyzes natural climate change, natural pandemics, near-Earth objects (asteroids, comets, and meteors), space weather (coronal mass ejections, solar flares, and solar particle events), stellar explosions (gamma-ray bursts and supernovae), and volcanic eruptions. Almost all natural GCR scenarios involve important interactions between the natural hazard and human civilization. Several natural GCR scenarios may have high ongoing probability. Deep human history provides little information about the resilience of modern global civilization to natural global catastrophes. The natural GCRs should not be dismissed on grounds of deep human history. Work on natural GCRs should account for their important human dimensions. A case can even be made for abandoning the distinction between natural and anthropogenic GCR.

Existential Risks to Humanity Should Concern International Policymakers and More Could Be Done in Considering Them at the International Governance Level

In this perspective, we consider the possible role of the United Nations (UN) with respect to existential risks to human civilization and the survival of humanity. We illustrate how existential risks have been discussed at an international governance level, specifically in documents in the UN Digital Library. In this large corpus, discussions of nuclear war account for over two-thirds (69%, 67/97) of mentions of existential risks, while mention of other existential risks, or such risks as a category, appears scant. We take these observations to imply inadequate attention to these significant threats. These deficits, combined with the need for a global response to many risks, suggest that UN member nations should urgently advocate for appropriate action at the UN to address threats, such as artificial intelligence, synthetic biology, geoengineering, and supervolcanic eruption, in analogous fashion to existing attempts to mitigate the threats from nuclear war or near-Earth objects.

An upper bound for the background rate of human extinction

We evaluate the total probability of human extinction from naturally occurring processes. Such processes include risks that are well characterized such as asteroid impacts and supervolcanic eruptions, as well as risks that remain unknown. Using only the information that Homo sapiens has existed at least 200,000 years, we conclude that the probability that humanity goes extinct from natural causes in any given year is almost guaranteed to be less than one in 14,000, and likely to be less than one in 87,000. Using the longer track record of survival for our entire genus Homo produces even tighter bounds, with an annual probability of natural extinction likely below one in 870,000. These bounds are unlikely to be affected by possible survivorship bias in the data, and are consistent with mammalian extinction rates, typical hominin species lifespans, the frequency of well-characterized risks, and the frequency of mass extinctions. No similar guarantee can be made for risks that our ancestors did not face, such as anthropogenic climate change or nuclear/biological warfare.

Making Evolutionary Sense of Gaia

The Gaia hypothesis in a strong and frequently criticized form assumes that global homeostatic mechanisms have evolved by natural selection favoring the maintenance of conditions suitable for life. Traditional neoDarwinists hold this to be impossible in theory. But the hypothesis does make sense if one treats the clade that comprises the biological component of Gaia as an individual and allows differential persistence - as well as differential reproduction - to be an outcome of evolution by natural selection. Recent developments in theoretical and experimental evolutionary biology may justify both maneuvers.

Questioning Estimates of Natural Pandemic Risk

The central argument in this article is that the probability of very large natural pandemics is more uncertain than either previous analyses or the historical record suggest. In public health and health security analyses, global catastrophic biological risks (GCBRs) have the potential to cause "sudden, extraordinary, widespread disaster," with "tens to hundreds of millions of fatalities." Recent analyses focusing on extreme events presume that the most extreme natural events are less likely than artificial sources of GCBRs and should receive proportionately less attention. These earlier analyses relied on an informal Bayesian analysis of naturally occurring GCBRs in the historical record and conclude that the near absence of such events demonstrates that they are rare. This ignores key uncertainties about both selection biases inherent in historical data and underlying causes of the nonstationary risk. The uncertainty is addressed here by first reconsidering the assumptions in earlier Bayesian analyses, then outlining a more complete analysis accounting for several previously omitted factors. Finally, relationships are suggested between available evidence and the uncertain question at hand, allowing more rigorous future estimates.

Individual Differences in Risk Perception of Artificial Intelligence

Abstract. This cross-sectional study (N = 325) investigated the relationship between the Dark Triad personality traits and the perception of artificial intelligence (AI) risk. Narrow AI risk perception was measured based on recently identified perceived risks in the public. Artificial general intelligence (AGI) risk perception was operationalized in terms of plausibility ratings and subjective probability estimates on deceptive AI scenarios developed by Bostrom (2014) , in which AI-sided deception is described as a function of intelligence. Machiavellianism and psychopathy predicted narrow AI risk perception above the shared variance of the Dark Triad and above the Big Five. In individuals with self-reported knowledge of machine learning, the Dark Triad traits were associated with AGI risk perception. This study provides evidence for the existence of substantial individual differences in the risk perception of AI.

Woodpeckers and Diamonds: Some Aspects of Evolutionary Convergence in Astrobiology

Jared Diamond's argument against extraterrestrial intelligence from evolutionary contingency is subjected to critical scrutiny. As with the earlier arguments of George Gaylord Simpson, it contains critical loopholes that lead to its unraveling. From the point of view of the contemporary debates about biological evolution, perhaps the most contentious aspect of such arguments is their atemporal and gradualist usage of the space of all possible biological forms (morphospace). Such usage enables the translation of the adaptive value of a trait into the probability of its evolving. This procedure, it is argued, is dangerously misleading. Contra Diamond, there are reasons to believe that convergence not only plays an important role in the history of life, but also profoundly improves the prospects for search for extraterrestrial intelligence success. Some further considerations about the role of observation selection effects and our scaling of complexity in the great debate about contingency and convergence are given. Taken together, these considerations militate against the pessimism of Diamond's conclusion, and suggest that the search for traces and manifestations of extraterrestrial intelligences is far from forlorn. Key Words: Astrobiology-Evolution-Contingency-Convergence-Complex life-SETI-Major evolutionary transitions-Selection effects-Jared Diamond. Astrobiology 18, 491-502.

Evolvability Is an Evolved Ability: The Coding Concept as the Arch-Unit of Natural Selection

Physical processes that characterize living matter are qualitatively distinct in that they involve encoding and transfer of specific types of information. Such information plays an active part in the control of events that are ultimately linked to the capacity of the system to persist and multiply. This algorithmicity of life is a key prerequisite for its Darwinian evolution, driven by natural selection acting upon stochastically arising variations of the encoded information. The concept of evolvability attempts to define the total capacity of a system to evolve new encoded traits under appropriate conditions, i.e., the accessible section of total morphological space. Since this is dependent on previously evolved regulatory networks that govern information flow in the system, evolvability itself may be regarded as an evolved ability. The way information is physically written, read and modified in living cells (the "coding concept") has not changed substantially during the whole history of the Earth's biosphere. This biosphere, be it alone or one of many, is, accordingly, itself a product of natural selection, since the overall evolvability conferred by its coding concept (nucleic acids as information carriers with the "rulebook of meanings" provided by codons, as well as all the subsystems that regulate various conditional information-reading modes) certainly played a key role in enabling this biosphere to survive up to the present, through alterations of planetary conditions, including at least five catastrophic events linked to major mass extinctions. We submit that, whatever the actual prebiotic physical and chemical processes may have been on our home planet, or may, in principle, occur at some time and place in the Universe, a particular coding concept, with its respective potential to give rise to a biosphere, or class of biospheres, of a certain evolvability, may itself be regarded as a unit (indeed the arch-unit) of natural selection.

Existential risks: exploring a robust risk reduction strategy

A small but growing number of studies have aimed to understand, assess and reduce existential risks, or risks that threaten the continued existence of mankind. However, most attention has been focused on known and tangible risks. This paper proposes a heuristic for reducing the risk of black swan extinction events. These events are, as the name suggests, stochastic and unforeseen when they happen. Decision theory based on a fixed model of possible outcomes cannot properly deal with this kind of event. Neither can probabilistic risk analysis. This paper will argue that the approach that is referred to as engineering safety could be applied to reducing the risk from black swan extinction events. It will also propose a conceptual sketch of how such a strategy may be implemented: isolated, self-sufficient, and continuously manned underground refuges. Some characteristics of such refuges are also described, in particular the psychosocial aspects. Furthermore, it is argued that this implementation of the engineering safety strategy safety barriers would be effective and plausible and could reduce the risk of an extinction event in a wide range of possible (known and unknown) scenarios. Considering the staggering opportunity cost of an existential catastrophe, such strategies ought to be explored more vigorously.

Evaluating methods for estimating existential risks

Researchers and commissions contend that the risk of human extinction is high, but none of these estimates have been based upon a rigorous methodology suitable for estimating existential risks. This article evaluates several methods that could be used to estimate the probability of human extinction. Traditional methods evaluated include: simple elicitation; whole evidence Bayesian; evidential reasoning using imprecise probabilities; and Bayesian networks. Three innovative methods are also considered: influence modeling based on environmental scans; simple elicitation using extinction scenarios as anchors; and computationally intensive possible-worlds modeling. Evaluation criteria include: level of effort required by the probability assessors; level of effort needed to implement the method; ability of each method to model the human extinction event; ability to incorporate scientific estimates of contributory events; transparency of the inputs and outputs; acceptability to the academic community (e.g., with respect to intellectual soundness, familiarity, verisimilitude); credibility and utility of the outputs of the method to the policy community; difficulty of communicating the method's processes and outputs to nonexperts; and accuracy in other contexts. The article concludes by recommending that researchers assess the risks of human extinction by combining these methods.

Small theories and large risks--is risk analysis relevant for epistemology?

Ought we to take seriously large risks predicted by "exotic" or improbable theories? We routinely assess risks on the basis or either common sense, or some developed theoretical framework based on the best available scientific explanations. Recently, there has been a substantial increase of interest in the low-probability "failure modes" of well-established theories, which can involve global catastrophic risks. However, here I wish to discuss a partially antithetical situation: alternative, low-probability ("small") scientific theories predicting catastrophic outcomes with large probability. I argue that there is an important methodological issue (determining what counts as the best available explanation in cases where the theories involved describe possibilities of extremely destructive global catastrophes), which has been neglected thus far. There is no simple answer to the correct method for dealing with high-probability high-stakes risks following from low-probability theories that still cannot be rejected outright, and much further work is required in this area. I further argue that cases like these are more numerous than usually assumed, for reasons including cognitive biases, sociological issues in science and the media image of science. If that is indeed so, it might lead to a greater weight of these cases in areas such as moral deliberation and policy making.