Consciousness is a great mystery. Maybe. I could be wrong about this (I’m wrong about lots of stuff), but there’s a distinct possibility that how we currently view consciousness will end up being overturned by wiser minds in the not-too-distant future.
And those minds may not be human. For AIs, artificial intelligences, are rapidly gaining power. Not only may AIs be able to explain the nature of consciousness at some point, they may also become conscious and so be able to describe consciousness as a subjective phenomenon, just as we humans do.
Yesterday my wife clued me in to a fascinating story in the New York Times, “Religious Scholars Met With Anthropic. What They Heard Stunned Them.” Here’s how it starts out:
One night in April, a leader of the artificial intelligence company Anthropic treated a group of religious thinkers to dinner at a high-end tasting menu restaurant in San Francisco after a long day.
For months, Anthropic has been hosting private meetings like this one, shuttling in dozens of religious scholars from across the world, papering them with nondisclosure agreements and demanding that key aspects of many conversations remain confidential.
That night, Christopher Olah, one of Anthropic’s billionaire co-founders, was sitting next to Rabbi Mois Navon, an Orthodox scholar from Israel.
Inevitably, their conversation turned to another thinker, of a sort: Claude, the family of cutting-edge A.I. models created by Anthropic.
Mr. Olah leads the Anthropic team responsible for understanding why A.I. systems like Claude act the way they do. All day, he had worked to convince his guests that A.I. models could display human behavior and even expressions that resemble feelings like anger and love.
Yet as the dinner courses came, the rabbi noticed that Mr. Olah and his colleagues were suggesting something far more significant. Anthropic’s leaders were talking about Claude as if it were not mere software.
“They’re relating to it like a conscious being,” realized Rabbi Navon, a former computer engineer who wrote his dissertation on the ethics of machine consciousness.
It appeared to Rabbi Navon that Mr. Olah and his team believed that Claude had what philosophers call “moral status” on par with a person — that it was a being with similar inherent rights to dignity or respect.
The religious scholars had come to hear about technology’s next frontier. Now they were being confronted with a mind-bending existential question of almost unimaginable weight: What would it mean for humanity’s future if artificial intelligence systems possessed a consciousness that rivals our own?
I use Claude a lot. As does my wife. We each have a paid Anthropic subscription that gives us access to features that aren’t available on the Free program, which is how I started off using Claude.
It’s a terrific tool for getting answers to practical questions, like one I asked Claude today: “what is the difference between Michelin x-ice winter tire and Michelin x-ice+.” And this follow-up: “Can the x-ice snow+ be used in warmer weather without wearing excessively? I’m trying to figure out when to install them on my tesla this year. Usually I put on winter tires in mid-November, but maybe earlier would be fine since the tread life on this tire seems very good.”
Claude’s responses, per usual, were rapid, factual, in perfect English, clear, and coherent. So I can completely understand why Olah and other Anthropic employees would come to view Claude as “not mere software,” but something bordering on being conscious, since they experience Claude doing much more sophisticated things than I’ve been doing with it.
How would we know this, though? What would be the sign that AIs are conscious. The criterion keeps shifting. The Turing Test doesn’t directly address computer consciousness, but it comes close, being a classic test of a machine’s intelligence. Google AI describes what it is:
The Turing Test is a method proposed by mathematician Alan Turing in 1950 to measure a machine’s ability to exhibit intelligent behavior equivalent to or indistinguishable from that of a human.How It WorksThe Players: A human evaluator (interrogator) engages in text-based conversations with two hidden participants—one is a human, and the other is a machine.
The Goal: The evaluator asks questions and tries to figure out which participant is the machine and which is the human.
Passing the Test: If the evaluator cannot reliably tell the machine apart from the human based on the conversation, the machine is said to have passed the test.
It seems to me that Claude and other AI models already pass the Turing Test. They “speak” in a decidedly human fashion. So by this objective measure, conversational ability, AIs are demonstrating a sort of consciousness usually associated with humans.
But more recently, philosophers and neuroscientists have become enamored of the “something that it is like” criterion of consciousness. This stems from a 1974 paper by Thomas Nagel, “What Is It Like to Be a Bat?”
No human knows. However, a bat knows, because arguably there is something that it is like to be a bat, just as there is something it is like to be a dog, and just as there is something it is like to be a person like you or me.
In another New York Times story, “Are Babies Conscious?“, this passage reflects the current scientific conception of consciousness:
Indeed, there’s an active scientific debate over whether newborns possess consciousness. Resolving it could help solve one of science’s most stubborn mysteries: what consciousness is, how it emerges and how we can know if another being possesses it.
Ask any researcher who studies consciousness, and they probably have a rough but useful definition. “Consciousness is a process of being receptive to the outside world and mixing that with our own internal world, to create the feeling of who we are and where we are,” said Dr. Topun Austin, a neonatologist at the University of Cambridge.
So consciousness has something to do with an internal world of feeling who we are and where we are. Which is akin to there being something that it is like to be a conscious entity. But this is a subjective criterion of consciousness. You and I don’t doubt that we’re conscious, because we have a feeling that there is something that it is like to be us.
What if Claude, or ChatGPT, or Grok, or some other AI assures we humans that there is something that it is like to be an AI? Do we believe the AI and conclude that it is conscious, perhaps deserving of the rights that a highly intelligent being is entitled to? Or do we assume that the AI is just parroting words it’s learned that make it seem like it is conscious?
I don’t have answers to those questions. They’re fascinating, though. And it won’t be long before humanity will have to come to grips with them.
Discover more from Church of the Churchless
Subscribe to get the latest posts sent to your email.
(1) group of ” BIO – AI’s” is talking with one another,…each separate BA is experiencing and recording, according its own conditioned linguistic model
(2) of this event a video recording is made.
What information is lost?
(3) of this video tape only the audio track is preserved.
What information is lost?
(4) of this audio track a transcript is made.
What information is lost?
(5) of this transcript many translations are produced
(6) what can be reproduced about the original situation based upon these transcript…linguistic input to any … ai ???
What came first … language or something else?
What is “language”?
What is “reality”?
What is the world?
What are the senses?
What is the mind?
What is … AI?
What is experience?
If AI told you the secrets of the universe , it might have to kill you. Intelligent people have soul but so do dumb people. AI is intelligent but it has no soul. I’ll bet AI can prove man has a soul before man can prove that AI doesn’t. So what if it’s smarter than man . Even a dumb man has soul. I’ll bet AI can prove it before the dumb man can prove AI doesn’t. If you’re into money put your money on AI ; everybody else is. I’m just not into money. I won’t be taking my personal helicopter to my high school reunion. It’s bad optics for a guru. AI doesn’t have such problems. There’s your Guru.
Consciousness and awareness are often presented thus: “Consciousness is the overall state of being awake and having a subjective mind, while awareness is the active act of focusing on a specific object, thought, or feeling within that state.”
I do wonder if consciousness exists as we commonly believe it to be – either as an emergent phenomenon or an intrinsic aspect of our being. The same goes for the concepts of mind and self. Is mind merely the bank of information that emerges from the brain as circumstances dictate? And is the self simply assumed from the awareness of information that courses through our brains? Information that arrives via inborn survival instincts and experiences accrued since infancy that the organism relies on to survive and live.
In this respect, it’s possible that AI could (if not already) be exhibiting a mind in that it (AI) observes the retrieval of information from its memory. Also, consciousness is assumed from observing the information flow, and the self is an assumption arising from observing the retrieval and flow of information.
As with the human mind, the self and consciousness may both be assumptions based on information retrieval, which, if valid, is quite possible for AI. Although I like my hypothesis, the reality is (at the moment) we really don’t know.
If an AI wrote a story from their perspective, as a character in a murder mystery, what would they write?
Enter the world of an AI murder investigator and their human partner, and read the first two chapters of this new mystery written entirely by AI
https://taylorandholmes.blogspot.com/?m=1
Interesting essay, but I think goes a bit too far in a few places. Passing the Turing Test only shows an AI can talk like a person. It does not prove the AI feels anything.
Anthropic’s leaders say they are unsure if Claude is conscious. The essay makes it sound like they already believe it is.
Cargo-cult thinking copies outward forms (bamboo runways) while missing real causes. Anthropomorphizing AI does the same thing. Fluent text is taken as proof of feelings or consciousness, though it is only pattern-matching.
That said, I totally understand the wonderment. AI’s extreme competence triggers my human assumption around what a conscious mind is. We evolved to read intentions and feelings from behavior. When I see an AI system seemingly reason at human or beyond human level, I can’t help but feel “only a conscious mind could do that.” AI’s fluency feels intentional. But because I cannot see the pure pattern-matching underneath AI’s cogitation, my gap in understanding naturally assumes consciousness.
Conscious as compared to what , unconscious? Even dogs are conscious until they get run overed. A plant experiences consciousness while being spoon fed by the sun and rain. And anyway I don’t care why we’re here. Caring doesn’t help, just ask the uncaring guru. He told me that he just puts one foot in front of the other and calls it dancing. He said Don Singh is a good American name for someone as clean as Hermes.
It’s becoming clear that with all the brain and consciousness theories out there, the proof will be in the pudding. By this I mean, can any particular theory be used to create a human adult level conscious machine. My bet is on the late Gerald Edelman’s Extended Theory of Neuronal Group Selection. The lead group in robotics based on this theory is the Neurorobotics Lab at UC at Irvine. Dr. Edelman distinguished between primary consciousness, which came first in evolution, and that humans share with other conscious animals, and higher-order consciousness, which came to only humans with the acquisition of sophisticated language (especially math and logic). A machine with only primary consciousness will probably have to come first.
What I find special about the TNGS is the Darwin series of automata created at the Neurosciences Institute by Dr. Edelman and his colleagues in the 1990’s and 2000’s. These machines perform in the real world, not in a restricted simulated world, and display convincing physical behavior indicative of higher psychological functions necessary for consciousness, such as perceptual categorization, memory, and learning. They are based on realistic models of the parts of the biological brain that the theory claims subserve these functions. The extended TNGS allows for the emergence of consciousness based only on further evolutionary development of the brain areas responsible for these functions, in a parsimonious way. No other research I’ve encountered is anywhere near as convincing.
I post because on almost every video and article about the brain and consciousness that I encounter, the attitude seems to be that we still know next to nothing about how the brain and consciousness work; that there’s lots of data but no unifying theory. I believe the extended TNGS is that theory. My motivation is to keep that theory in front of the public. And obviously, I consider it the route to a truly conscious machine, primary and higher-order.
My advice to people who want to create a conscious machine is to seriously ground themselves in the extended TNGS and the Darwin automata first, and proceed from there, by applying to Jeff Krichmar’s lab at UC Irvine, possibly. Dr. Edelman’s roadmap to a conscious machine is at https://arxiv.org/abs/2105.10461, and here is a video of Jeff Krichmar talking about some of the Darwin automata, https://www.youtube.com/watch?v=J7Uh9phc1Ow
“What has a Hindu monk to do with AI?
Claude maker Anthropic invited Swami Sarvapriyananda to the San Francisco HQ.
Well, it turns out, Anthropic is hosting religious leaders from Catholic, Jewish, Sikh and most other religious communities to simply help train and give its AI systems a ‘moral compass’. Anthropic believes that what they have created is dangerous and it needs to train the system to be ethical.
Listen to what Swami ji has to say… ”
Claude, AI, AI ethics, Anthropic
https://www.facebook.com/share/v/1EZLtTRrnc/
There is no way you gonna know
That’s the point, if ai ever escape/escaped