** Statement:** The challenges for informatics scientists to confer “Sentience” in machine responses(AI computing) are many.
There are scattered benchmarks on the significant thresholds upon which a human observer’s scientific belief establishes that “the machine has become aware of its phenomenal consciousness”, beyond the artificially constructed phenomenal judgment in the machine, executed using a simple I/O computing apparatus, alongside a temporal learning model.
Objective: To benchmark a Sentient hypothesis or a degree of Sentience in an AI machine.
Ref: Sentience in an intelligent (linguistic driven) chat system of Google
Ref: As Chalmers puts it (1996, p.174):
“Judgments can be understood as what I and my zombie twin have in common. A zombie(Z) does not have any conscious experience, but “Z” claims that “Z” does. My Z twin judges that Z has conscious experience, and Z judgments in this vicinity correspond one-to-one to mine. Z will have the same form, and Z will function in the same way in directing behavior as mine… Alongside every conscious experience there is a content-bearing cognitive state. It is roughly information that is accessible to the cognitive system, available for verbal report, and so on.”
This Open Research is divided into few chapters. Please engage in community discussions on each chapter specific research.
Ch- 1: Deploy a case study of a simple test AI computing unit (Z’), which has three-factor interaction with Z’ surroundings.
Ch- 2: Identify qualias of an assumed state of phenomenal conscious experience of Z’, drawn from its three-factor interactions.
Ch-3: Observe the Paradox of Phenomenal Judgment, drawing conclusions from the Phenomenal Judgments of Z’.
Open Research Title: “Paradox of Phenomenal Judgment” in AI systems
- This is a multi-disciplinary Open Research. Please practice Foundation’s guidelines on the defining epistemic justice in the conventions used.
Open source, inferences on qualia properties of phenomenal consciousness
- *Keywords: * sentient AI, sentience in AI
Read Research Guidelines.