The Adjacent Possible

In this article, the authors propose intelligence is creating opportunities for a rich palette of future histories or what I have labeled 'the adjacent possible' in previous posts. That is, certain behaviors ("let's see if we can stand this post upright.") increase entropy and create future opportunities to increase entropy on top of entropy. These are dissipative structures -- energy is required to construct the structure and marginal energy added 'on top' of lower levels. The total entropy of a hierarchy of DS layers is bringing richer (higher entropy) futures into higher probability reach.

Increasing Entropy = Intelligence

"Entropica's intelligent behavior emerges from the "physical process of trying to capture as many future histories as possible," said Wissner-Gross. Future histories represent the complete set of possible future outcomes available to a system at any given moment.

Wissner-Gross calls the concept at the center of the research "causal entropic forces." These forces are the motivation for intelligent behavior. They encourage a system to preserve as many future histories as possible.

Reaching back to Roko's Basilisk, upright walking (above) creates the necessary precursor behaviors for the emergence, in time and with additional emergence(s) of writing code, tuning hardware and the other ancillary functions needed to develop the Basilisk. In this sense, developing a detailed plan to design and launch a product in the future causes actions in the present (and yesterday afternoon) that have causal links 'from' the desired future state. As the probability of the future state (a product for sale) breeds more and more positive behavior 'for' that outcome (weekly status updates, remediation activity for slipping schedules,..., usability documents) the desired outcome moves from thoughts to action (the CPU is integrated with the networking engines in an SoC manufactured in a target process).

In a sense, design teams share a persistent delusion -- the future is knowable -- that causes present action. Intelligence creates high entropy conditions from which something novel results. The new, new thing is one, two or more holonic layers above just standing upright and walking. But standing upright causes lots of other events besides high end network processing machines. Standing upright also causes web browsing and locating family members with smart phones. And some of the people finding family members on smart phones are also caused to work on network processors by a shared delusion that the future is amenable to present behavior.

The distinction between doing things totally in the present (eating lunch) and doing things which seem to provide necessary causes for far future things (improving Google search) is mundane. Imagine that I want to travel to France in the future and that I begin taking steps today so that my future self can see Paris in the spring. I enrole in a conversational French class, apply for my passport and begin to save for the trip. My present behavior is caused by an imaginary event in the future. I can monitor my behaviors most likely to cause the future trip with Bayes Theorem. My present behaviors result in either greater or lesser likelyhood the trip to France. Saving money and speaking French make the many possible futures regarding France more (or less) possible. If I have a passport, money for my flight and fluency the future with France is adjacent to a future self that can, with little effort, travel to France. The future where I travel to China is also an adjacent possible since I have a passport and money for travel there.