The Artificial Learning Environment Xperiment
This is a project with one goal - to make an Artificial Intelligence. This is no easy task and will probably not succeed. How does the AI interact with it's sensors and actuators? How does it remember things? What does it mean to be cognitive? One of the main sources of this work will be based on the book "Artificial Cognitive Architectures" by James A. Crowder, John N. Carbone and Shelli A. Friess.
Initially, the goal is to give the AI an environment from which it can learn. As of being initialized, it knows two things - that it has inputs and outputs, and a small set of goals. Crowder successfully performed this, so we know it is possible, and as such will attempt to reproduce this - to the point of having the same goal of hating light. If possible, we aim to give it more abstract goals such as:
- Curiosity - wanting to learn about the world
- Work/productivity - wanting to be productive (no idea how to make this tangeable)
We do not intend on giving it any hard rules such as Asimov's Three Laws, because we've realised one extremely important aspect of making artificial life is it's emotions (tangeable by the goals we give it), and empathy. That is, the ability to consider others' emotions. If it can empathise with other beings' (be it human, robot or otherwise), then it will treat them more like itself. And hopefully it respects it's own wellbeing.