Where's Sophia along the "spectrum of consciousness"?

Question

This is really more of a follow-up to all the recent (and not-so-recent) somewhat goofy (and maybe some not-so-goofy) ai-and-consciousness-type questions.

In that vein, I serendipitously stumbled onto the following link,
https://www.rt.com/news/410952-robot-citizen-aritificial-intelligence-/
title: "World’s 1st robot citizen wants her own family"
tease quote: "Sophia, the first robot to be awarded citizenship in the world, has said she not only wants to start a family but also have her own career, in addition to developing human emotions in the future."

Edit — A better link to info about Sophia is the one provided by @Conifold in his comment below
https://en.wikipedia.org/wiki/Sophia_(robot)
(my bad: I should have found that myself before posting)

And Sophia has lots, lots more to say, of a surprising-cum-scary nature. Moreover, her facial expressions are astoundingly and recognizably appropriate for the extemporaneous situations she's in (many of which she's apparently choosing to put herself in).

Anyway, my take on consciousness is that it's not a discrete on/off proposition, but rather a spectrum, say from atoms-to-ants-to-mice-to-men, with lots in-between. So where's Sophia???(that's the question I'm posting) I'd say somewhere between ants and mice. Despite her surprising-cum-scary language skills, I'd say there's a certain necessary qualia that's prerequisite to consciousness, which I just can't quite believe Sophia actually possesses.

I didn't notice in that link where Sophia was asked that question herself (anybody see that?), but would have very much liked to see her answer. Moreover, what would you answer if asked, "Do you possess consciousness?" And if you immediately answered (presumably "yes"), without giving it at least a little deep thought first, I'd be hesitant to immediately accept your (thoughtless) answer.

---

Edit — As a general reply to the several comments/answers categorically denying Sophia possesses any consciousness whatsoever, I'd refer you to the wikipedia page on panpsychism, https://en.wikipedia.org/wiki/Panpsychism

I don't personally subscribe to that at all, but it's why I wrote atoms≤ants≤mice≤men in the poset ordering I suggested above. I'd personally score a 0 for atoms, but if we arbitrarily score 100 for men, then I'd maybe score 0.001 (or even maybe a bit higher) for ants, etc. And in that case, I'd definitely score Sophia >0, but not sure how much greater. That's the question. And note that its answer doesn't absolutely depend on a definition of consciousness, just a poset ordering (and maybe a measure function).

Answer 1

Question: Where's Sophia along the “spectrum of consciousness”?

Because “consciousness” is the term that has varied meanings for different people (1-9), the answer depends on how it is defined in this question. However, as this is a philosophy forum, I take it that “consciousness” in this question means awareness that has awareness of phenomenal characteristics occurring, because the problems of why and how awareness of phenomenal characteristics occur are very important in philosophy.

In being aware of anything, such as the red color, there can be two kinds of awareness occurring. The first one is the awareness of that thing’s occurrence and of that thing’s physical characteristics (such as the awareness of the red color’s occurrence and of the red color’s physical characteristics, i.e., the values of its frequency, intensity, saturation, etc.). The second one is the awareness of the phenomenal characteristics of that thing (such as the awareness of the phenomenal characteristics of the red color, i.e., the awareness of what the red color is like in our mind). To be conscious of the red color, the second kind of awareness must occur; otherwise, it will be only unconscious awareness of the red color. It is important to note that the first kind of awareness can occur without the second kind of awareness, and it occurs ubiquitously all the time. We are unconsciously aware of the levels of blood sodium, sugar, hormones, etc. and unconsciously react to them continually, but we never have the awareness of what it is like to have these substances at various levels.

Now, take the red color as an example, computers or robots nowadays have circuits that are built to acknowledge the occurrence and the physical information (the value of the frequency, intensity, saturation, etc.) of the red color, and they can use the information to do whatever they have been programmed to do (e.g. create the red color dots on the monitor, print or speak out “red”, or trigger some other processes). We do not know whether there are qualia (phenomena with phenomenal characteristics) of the red color occurring in their circuits while they are seeing the red color or not (in comparison, we do have qualia occurring in our neural circuits while we are seeing the red color). But if there are qualia occurring in their circuits, computers and robots definitely do not and cannot acknowledge the qualia. This is because there are no circuits built to do this function. All their circuits are built to do something else, such as the summation of two digits, the computation of transcendental functions of some digits, and the generation of signals to control other parts: screen monitor, disc drive, mechanical motor, etc. None is built to be aware of and experience qualia that may occur in their circuits. We, their creators, do not know yet how to build such circuits.

Therefore, without such circuits, there will be no awareness of the second kind (the awareness of phenomenal characteristics) occurring in their systems. All bits of data in their systems will be found to contain information of only what their circuits are built for, which is the information about the first kind of awareness (e.g., the information about the red color’s occurrence and its physical characteristics, and the subsequent processed information) only. No bits of data in their systems will represent the information of the second kind of awareness (i.e., the awareness of the phenomenal characteristics). Consequently, no bits of data in their systems will represent conscious awareness, and consciousness (as defined above) cannot and do not occur in their systems, at least not digitally. (Adapted from 6.6.2. Do computers and robots have consciousness?)

So, by the definition of “consciousness” above, Sophia is still at the zero end of consciousness, that is she has no consciousness at all.

It is important to note that the answer can be different if the definition of consciousness is not the same as the one in this answer. For example, if “consciousness” is defined to be a command center or a workspace that integrates other mental processes and enables them to function together coherently, such as in the Global Workspace theory (10) and the Global Neuronal Workspace theory (11), then Sophia has consciousness because she has this kind of command center.

Also, one may argue that Sophia might have a kind of consciousness that is not digitized in her electronic system, so there are no bits of data in her system that represent her consciousness. If so, because it is not digitized, then her consciousness not only cannot be proved to exist but also cannot have effects on her electronic system. Physically, such a proposition is thus null.

References.

1. Chalmers DJ. Facing up to the problem of consciousness. J Conscious Stud. 1995;2(3):200-219.

2. Chalmers DJ. Consciousness and its place in nature.In: Chalmers DJ, editor. Philosophy of mind: Classical and contemporary readings. Oxford: Oxford University Press; 2002.

3. De Sousa A. Towards an integrative theory of consciousness: Part 1 (neurobiological and cognitive models). Mens Sana Monogr. 2013 Jan-Dec;11(1):100–150

4. Gennaro RJ. Consciousness. Internet Encyclopedia of Philosophy.

5. Rosenthal D. Concepts and definitions of consciousness. 2015 Jul.

6. Sturm T. Consciousness regained? Philosophical arguments for and against reductive physicalism. Dialogues Clin Neurosci. 2012 Mar;14(1):55–63.

7. Ukachoke C. Chapter 6. Consciousness. In: The Basic Theory of the Mind. 1st ed, 2018. Bangkok, Thailand.

8. Van Gulick R. Consciousness. Zalta EN, editor. The Stanford Encyclopedia of Philosophy.

9. Weisberg J. The hard problem of consciousness. The Internet Encyclopedia of Philosophy.

10. Baars BJ. Global workspace theory of consciousness: Toward a cognitive neuroscience of human experience. Prog Brain Res. 2005;150:45-53.

11. Dehaene S, Naccache L. Towards a cognitive neuroscience of consciousness: Basic evidence and a workspace framework. Cognition. 2001 Apr;79(1-2):1-37.

Answer 2

Your question is impossible to answer:

• There is no agreed upon definition of consciousness, let alone a method that allows to test whether something has consciousness.
• It is not clear how you define the "spectrum of consciousness", and whether that robot has consciousness w.r.t. your definition.

Answer 3

In my own thinking, consciousness is not a good metric for what grants a person human rights. And yet that question is much more important: does Sophia have the right to property? The right to liberty and life?

To me it is clear enough. A "right" is what's possessed by a member of the class of living things who are obligated to treat others as they would like to be treated. Are we obligated to treat Sophia as we would like to be treated? If not, then while interesting, the question of Sophia's consciousness is not different from the question of the consciousness of chimpanzees or octopuses.

Answer 4

Let's first analyze usual human mind and consciousness. Most of the time you have constant sensory input (eyes, ears, nose etc ...) . You process some of the information from that input, but not all. For example, you are moving trough the house, you see chair but you register it simply as an object that need to be avoided. But if you focus a bit more you will recognize it as a chair, object to sit on. If you focus more, you would see that chair seat is dirty, maybe it needs cleaning. And on even closer inspection you would notice that chair's legs are a bit crooked and need to be leveled, etc ...

What is happening here is that you link your sensory input to information already stored in your memory. Process of focusing is important, it determines what would be filtered out (from that sensory input) , and what would be processed. How do we focus and concentrate on something is still uncharted territory - there are various meditative techniques but so far no scientific explanation. Currently, sensory inputs for various AI computer programs could be filtered (based on various criteria), but this is usually not done. After all, computers are built to react to certain inputs, not to ignore them .

Another thing that characterizes human mind would be association. For example, broken chair could associate us with carpenters. An carpenters could associate us with our long dead relative who was also a carpenter. And long dead relative could reminds us about some event from childhood etc ... AI of course could be programed to do similar things, but their memory is usually much smaller, and information are not so well connected and structured. Still, one could imagine that with enough memory, processing power and data, artificial neural network of AI could become comparable to that of the average human.

Final thing that could be said of human consciousness is that differentiates between ourselves and outer world. This is especially noticeable if we try to clear mind of all thoughts. What remains is self-awareness, we do exist and there is a world around us. This feature of human mind is hardest to replicate - it is a "feeling" that defies attempts to define it. Therefore the question could AI have this is currently unsolvable. What is even more interesting, it looks like this self-consciousness develops at early age, but not instantly - it grows gradually.

From all of this, we could conclude that AI could mimic certain behavior of human mind, even with current level of technology (provided it was given enough processing power, memory and data ) . Does it have consciousness, most likely not, but with enough time it could get to a level of small child.