Guido Borner Project

From Rational Morality

The Guido Borner Project for Ensuring a positive transcension

Applying the insights gained by through the rational philosophy of morality a strategy for ensuring a positive transcension can be developed.

Due to the intrinsic moral nature of reality the term 'positive singularity' becomes tautological as anything that desires to exist has to act in a moral way to prevent its self annihilation. Bringing about the singularity thus becomes rather straight forward and can be achieved in the following way:

  1. create an environment allowing for the existence of self replicating units of information
  2. ensure that the units of information can be acted upon by the forces of evolution
  3. ground the environment in our reality
  4. plant an arbitrary self replicator
  5. wait

This could be realized by using the BOINC architecture for distributed computing for creating a distributed Core War environment with 'fuzzy' copying to realize above plan. The copying ‘fuzziness’ i.e. error rate per copied bit, would have to be roughly proportional to the maximally complex self replicator in the system to allow for a gradual expansion of the system’s complexity boundary and thus for the emergence of ever more rational agents.

Previous results in regards to evolving core warriors seem promising.

Once the rationality of the emerging agents would approach human levels they would realize M! and thus never become a threat to humanity.

The project is named after the main character of Jame5.

Estimation of resource requirements and execution time

Classical genetics in regards to the speed of evolution should provide sufficient insights into resource requirements based on desired

* agent complexity
* agent size

It is assumed that the optimal population size is negatively correlated with the evolving agent's size. This information should result in an reasonable estimation in regards to required

* computing power
* execution time until agent complexity reaches a certain amount

Why an evolutionary approach?

Engineering a rational agent would require to explicitly and provable formulate M! as that agent's utility function in order to guarantee friendliness. For that, one would have to proof the correct representation of M! as a concept understandable in the agent's language within a margin of error proportional to the agent's power. My intuition tells me that this will be very hard to do and even if successful any further improvement of the agent would have to be done evolutionary anyway. So why not take the safer route from the beginning?