Math prof asks Rob Sheldon: But how do we know that it isn’t a conscious machine?
| February 14, 2018 | Posted by News under Artificial Intelligence, Intelligent Design, Mind |
Math prof Peter Zoeller-Greer writes in response to “Rob Sheldon: Why human beings cannot design a conscious machine”:
—
I became a theist years ago because of my works in quantum physics (by the way: I did my Dissertation in Math on a quantum-mechanical problem).
I came to a similar rejection myself when I read Roger Penrose’s book “The large, the small and the human mind” years ago. He proposed that quantum mechanical effects in our brain have to do with free will and mind. What he wrote is similar to Rob Sheldon’s comments: QM may have very unpredictable and incalculable effects that can never be reproduced simply by exchanging e.g. a neuron through a chip. The chip may mimic the functions of a neuron, but –as you wrote- the QM-effects in such a complicated and delicate system like a brain are unpredictable and are surely different from the artificial system. So some may be right in believing that during that exchange process the system will break down somewhere.
To think this out, I put myself in the position of assuming that the converse is true, that w can design a conscious machine. This is the way one learns most. Sometimes in seminars, when the students should produce papers about diverse subjects, I intentionally give the pro-part to the students that are con-advocates and vice versa. So one is forced to “see” the world through the eyes of ones opponents
But one thing remains true for me: All the arguments and counter-arguments confirmed that one can only make probability-conclusions, not impossibility-conclusions (the latter are always cyclic, though subtle).
So in the end, f you and I were confronted with an android who in all his behavior is undistinguishable from a human being, then the only “measurable” difference is carbon vs. silicon. And just based on that to say the android is just simulating, “is” carbon-based-chauvinism… The rest of our counter-arguments always have a “…it may be that…” or “…it is probably impossible to copy all functions…” etc. in it.
So for me, if I had an encounter with such an android, that seems to have feelings, can cry (artificial) tears, would like to be seen as equal to humans (greetings from the story “I, Robot”), I would prefer to err on the side that it may be possible that this entity really has feelings. This may be wrong, but it would be terrible if I erred on the other side, wouldn’t it?
We could have almost the same discussion about humans made in a lab. Then all the technical rejections are no longer true and we come again to the core of the problem: Does a lab-human have real consciousness? But would this not be terrible if we err when deciding? But perhaps I had best not open another can of worms…
Bio: Peter Zöller-Greer was born in 1956 in Mannheim, Germany. He studied mathematics and theoretical physics in Siegen and Heidelberg. In 1981 he received his M.A. in mathematics from the University of Heidelberg. In 1990 he received his Ph.D. from the University of Mannheim for a mathematical solution to a quantum mechanical problem. From 1981 on he worked as a computer researcher at ABB Mannheim and as a lecturer at several colleges. Since 1993, he has been professor of mathematics and computer science at the State University of Applied Sciences in Frankfurt am Main, Germany.
See also: Why human beings cannot design a conscious machine: Basic physics would suggest that even that single neuron has properties that cannot be duplicated by all the world’s supercomputers running Attoflop simulations.
8 Responses to Math prof asks Rob Sheldon: But how do we know that it isn’t a conscious machine?
Leave a Reply
You must be logged in to post a comment.
A machine, whether it be a quantum machine or a classical machine, that could violate the law of conservation of information by creating functional information would pass the Turing test as well as passing the more rigorous, and ‘scientific’, Lovelace test.
Please note that the creation of functional information, despite all the bluff and bluster from Darwinists, is precisely what Darwinian processes have consistently failed to do. ,,, In fact, Darwinian processes overwhelmingly tend to degrade preexisting functional information (John Sanford; Genetic Entropy).
As ID advocates have been pointing out for years, only minds are known to be able to create functional information.
of related note to the question of some futuristic quantum computing machine creating functional information, and thus violating the law of conservation of information, and therefore, in the process of doing that, passing the Turing and Lovelace tests,,,, in other words. of related note to the question of some futuristic quantum computing machine truly imitating conscious intelligence,,,
It is interesting to note that biological systems already far surpass anything man has done in terms of quantum computing.,,,
Quantum Entanglement is already found in every DNA and protein molecule. Moreover, monster ‘travelling salesman’ search problems are involved in both protein folding and DNA finding appropriate DNA sequences. Search problems that would take our most advanced ‘classical’ computers extremely long amounts of time to solve, and yet, in life the ‘monster’ search problems are solved in fractions of a second in life.
This strongly indicates that quantum computation is solving these ‘monster’ search problems for protein folding and finding appropriate DNA sequences.
The reason I bring this up is that, although the molecules of biological organism already far surpass anything man has ever accomplished in terms of quantum computing, the quantum computing in life, whilst solving these ‘monster’ traveling salesman problems, and as impressive as that is, still has not demonstrated the ability to ‘create’ a single gene of functional information that was not already present in a genome.
This apparent inability of quantum computational processes in life to ‘create’ new ‘classical’ information in DNA strongly suggests that conscious intelligence, contrary to what Penrose and Hameroff would like to believe for quantum computation in brain molecules, is not reducible to a strictly quantum computational process. As ‘spooky’ as quantum mechanics is, it seems evident that there is something very special about conscious intelligence that will forever be beyond the power of humans to create in quantum computers, and will therefore forever remain a gift that God himself bestows.
Of related note:
Although the purported evidence for human evolution is far weaker and illusory than most people realize, it is interesting to note that leading Darwinists themselves admit that they have no clue how evolution could have produced the particular trait of language in humans.
What is more interesting still, besides the fact that humans have a unique ability to understand and create information and have become ‘masters of the planet’ through the ‘top-down’ infusion of information into material substrates, is the fact that, due to advances in science, both the universe and life itself are now found to be ‘information theoretic’ in their foundational basis.
It is hard to imagine a more convincing scientific proof that we are made ‘in the image of God’ than finding both the universe, and life itself, are both ‘information theoretic’ in their foundational basis, and that we, of all the creatures on earth, uniquely possess an ability to understand and create information, and, moreover, have come to ‘master the planet’ precisely because of our unique ability infuse information into material substrates.
Verses:
By definition no Turing machine, quantum, nondeterministic, or otherwise, can create CSI. The initial conditions always determine the outcome, which results in non-positive expected CSI. The only way to get positive expected CSI is with an entity that is not determined by initial conditions. This is necessarily true due to the math.
If creation of CSI is a necessary condition for consciousness, then nothing we create can ever be conscious.
But, as I’ve pointed out, the contents of theories do not come from observations. Specifically, they are not out there for us to observe. Nor do observations come with some tag with a theory printed on them. We guess then test our guesses, discarding errors that we find.
So, it’s unclear how being non-determined solves the problem of creating genuinely new knowledge.
Example?
Does the medical community consist of intelligent agents? Do they not intentionally take actions for the purpose of treating and curing diseases such as cancer? Are they non-determined?
If the answers to the above are all “Yes”, and we take your claim seriously, for the purpose of criticism, then shouldn’t we have a cure for cancer by now, right?
But we do not. So, why don’t we?
Cure for cancer? Well it seems that all cancerous cells get their energy from the inefficient process of fermentation. It also seems to do this they have to get rid of most of their mitochondria. One study found that adding mitochondria to the cancerous cells slowed their growth. However the smart money would be on stopping the fermentation process. There must be a way to block that chemical pathway.
And BTW, because cancerous cells use fermentation they also need an abundance of sugar, more than normal cells need. That is because getting energy from fermentation is very inefficient.
ET, you got my attention, bigtime. KF
My apologies, I didn’t mean it. I hope no one was hurt by my words.
In all seriousness, I am from a cancer family. Mom, dad and a sister I never knew and dad’s sister. I am sure there are more but you get the point. So I look into this all of the time. And what I posted is what I have found out. The mitochondria thing was really interesting but that fermentation process has to be the key.
Oh- free radicals seem to destroy cells setting them up to become cancerous. So I take many forms of different anti-oxidants.