Technically Exists

Singletons and universal inevitable threats

2018-02-09

One concept I find particularly fascinating is that of the singleton. As defined by Nick Bostrom, a singleton is “a world order in which there is a single decision-making agency at the highest level”. He states that such an agency would have “the ability to prevent any threats (internal or external) to its own existence and supremacy”. However, it is not at all obvious to me that this is true. To crystallize my intuition I would like to introduce my own concept: the universal inevitable threat.

An inevitable threat is defined as a threat that eliminates an agency regardless of the actions that agency takes. This leaves open the possibility of delaying the threat for a finite period of time, but beyond that, the agency may only choose how to spend the time it has prior to the threat’s realization. It can neither eliminate the threat nor hold it off indefinitely.

A universal inevitable threat is then defined as an inevitable threat that applies to all possible agencies. A notable property of such a threat is that it cannot be brought about by an agency, as this would imply that the agency had a course of action that would result in the threat not occurring and therefore that the threat was not inevitable relative to that agency. This concept is relevant because if a universal inevitable threat exists, then no agency is able to prevent all threats to itself, including a singleton. Furthermore, several plausible candidates for a universal inevitable threat can be identified.

The first possibility is simply the end of the universe. Exactly how the universe will end has yet to be determined, but most hypotheses posit scenarios that appear to be universal inevitable threats. One example is the heat death of the universe, which would result in a state of maximum entropy in which no information processing can take place. Since information processing seems to be essential for agencies to operate, this scenario would count as a universal inevitable threat.

Another potential end of the universe is the Big Rip scenario. In this case, the universe’s increasing expansion rips everything apart into elementary particles and radiation. Unless an agency can somehow develop a way to survive the infinite expansion rate that is eventually reached, all agencies are eliminated, making this a universal inevitable threat.

The Big Crunch is yet another possible end state of the universe. This possibility involves the universe collapsing into a dimensionless singularity. A related scenario is the Big Bounce, which follows the Big Crunch with another Big Bang to create a cycle. Unless some agency can survive a singularity of infinite density, both cases are universal inevitable threats.

Another category of universal inevitable threat candidates is resource depletion scenarios. Some scenarios for the end of the universe can actually be framed as falling under this category. For example, heat death can be viewed as the universe running out of negentropy. However, it is possible that some other resource is both essential for agencies to function and depleted prior to the end of the universe. The depletion of such a resource would be a universal inevitable threat.

The final category of candidates for universal inevitable threats is that of naturally occurring physics disasters. An example of this would be false vacuum decay. If the universe is currently part of a metastable vacuum, it could be disrupted, creating a bubble of lower-energy vacuum that expands at the speed of light. This could fundamentally alter the universe in such a way as to destroy any agencies in existence prior to the event. False vacuum decay and any similar physics disasters that occur naturally would likely be universal inevitable threats.

While it is not yet possible to confirm the existence of a universal inevitable threat, there does seem to be a significant probability of one existing. In particular, the end of the universe is very likely to be one. If one exists, it would prove that no agency is able to prevent all threats to itself, and thus that singletons do not possess this power, as seems probable to me.


comments powered by Disqus