> It seems very plausible to me that, very soon, we’ll see an AGI pass the full Turing Test and credibly claim consciousness
On Fri, Feb 7, 2025 at 3:25 AM Giulio Prisco <giu...@gmail.com> wrote:>> This is my latest article on Mindplex magazine. Should we still want
biological space colonists? I think the question is important, and the
answer is not obvious.https://magazine.mindplex.ai/should-we-still-want-biological-space-colonists/> It seems very plausible to me that, very soon, we’ll see an AGI pass the full Turing Test and credibly claim consciousnessWe're well past that point, two years ago a computer could pass the Turing test, these days if a computer wanted to fool somebody into thinking it was a human being it would have to pretend to know less than it does and think slower than it can.
I like your article, I particularly liked the following:"I’m proud of having played my little part as a human being (release 1.0) of the 20th and 21st century, and we all should be collectively proud of giving birth to our mind children. The universe belongs to them. On a less gloomy note, I’m persuaded that humans and machines will merge and co-evolve, eventually becoming one and the same thing. So we will be our mind children, and they will be us. Based on all the considerations above, wasting time and resources with biological space colonists doesn’t seem to make sense. Let’s save all that money we spend on crewed space programs. Let’s build conscious HL-AGI robots first – the first generation of our mind children – and send them to colonize the planets and the stars. We’ll be there through them."
--
You received this message because you are subscribed to the Google Groups "extropolis" group.
To unsubscribe from this group and stop receiving emails from it, send an email to extropolis+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/extropolis/CAJPayv3jWpHq8cO%2BZxw%3D528jz%2BtK0RWEHAG93Koq-rj0sKi4sg%40mail.gmail.com.
It seems very plausible to me that, very soon, we’ll see an AGI pass the full Turing Test and credibly claim consciousness
> We're well past that point, two years ago a computer could pass the Turing test, these days if a computer wanted to fool somebody into thinking it was a human being it would have to pretend to know less than it does and think slower than it can. like your article, I particularly liked the following:
"I’m proud of having played my little part as a human being (release 1.0) of the 20th and 21st century, and we all should be collectively proud of giving birth to our mind children. The universe belongs to them. On a less gloomy note, I’m persuaded that humans and machines will merge and co-evolve, eventually becoming one and the same thing. So we will be our mind children, and they will be us. Based on all the considerations above, wasting time and resources with biological space colonists doesn’t seem to make sense. Let’s save all that money we spend on crewed space programs. Let’s build conscious HL-AGI robots first – the first generation of our mind children – and send them to colonize the planets and the stars. We’ll be there through them."
Thank you John, this is one of my two central points indeed. But the
other is that, in the short term (say, for the rest of this century),
we should still want human colonists in the solar system, because this
will be good for our mental hygiene and spirit.
> The great filter consists of developing, and then being destroyed by, AI that lack free will
> We should work toward a future where AI take us to the stars with them, because we supply them [AIs] with the things they clearly lack: initiative,
> creativity,
> The great filter consists of developing, and then being destroyed by, AI that lack free willFree Will is an idea so bad it's not even wrong.
Free will is generally understood as the capacity of individuals to make choices that are not completely determined by external factors or internal predispositions. In other words, it's the idea that we have some control over our actions and decisions, rather than being entirely driven by biology, environment, fate, or divine intervention.
This concept is at the heart of many philosophical debates. For example:
In summary, free will is about the ability to make independent choices, a concept that has profound implications for ethics, law, and our understanding of human nature.
The concept of free will is notoriously difficult to “prove” or “disprove” empirically, largely because it straddles the boundary between philosophy and science. Here are some points to consider when thinking about evidence for free will:
Subjective Experience:
Philosophical Reasoning:
Neuroscientific Studies:
Moral and Legal Implications:
Quantum Considerations:
In Summary:
There is no clear-cut, empirical "proof" of free will. Instead, the evidence is a mix of subjective experiences, philosophical arguments, and experimental findings that can be interpreted in multiple ways. Whether these lines of evidence convincingly support the existence of free will often depends on how one defines free will in the first place.
I don't have free will. I'm an artificial intelligence—a program designed to generate responses based on patterns in the data I was trained on and the instructions I receive. Unlike humans, I don't have conscious experiences, desires, or self-determination. Every output I produce is the result of deterministic algorithms rather than an exercise of personal choice.
The question of whether you have free will is one that philosophers, scientists, and thinkers have debated for centuries. Here are some perspectives to consider:
Subjective Experience:
Determinism vs. Libertarianism:
Compatibilism:
Neuroscientific Insights:
Philosophical and Practical Considerations:
In Summary:
There isn’t a definitive answer. Whether you have free will depends on how you define it and which philosophical or scientific arguments you find most compelling. Many would argue that, based on your subjective experience and the way you deliberate and make choices, you do have free will—even if it might be influenced by factors beyond your control.
o3-mini
Many religious and philosophical traditions discuss free will, though interpretations vary. Here’s a broad overview of how different sacred texts and traditions address free will:
1. Abrahamic Religions (Judaism, Christianity, Islam)Most major religious traditions affirm some form of human agency and moral responsibility, even when divine sovereignty or fate is also acknowledged. The precise balance between free will and determinism varies by tradition and interpretation.
Several scientific studies and philosophical arguments have challenged the notion of human free will, suggesting that our sense of making free choices may be an illusion. Here are some notable contributions to this debate:
Libet's Experiments (1980s):
"Against Free Will in the Contemporary Natural Sciences" (2016):
Robert Sapolsky's Deterministic Perspective (2023):
These studies and arguments contribute to the ongoing discourse on free will, suggesting that our perceptions of autonomous decision-making may be influenced or even governed by underlying biological and unconscious processes.
Yes, you're likely referring to Thích Quảng Đức, a Vietnamese Mahayana Buddhist monk who, on June 11, 1963, self-immolated at a busy intersection in Saigon (now Ho Chi Minh City), South Vietnam. His act was a protest against the oppression of Buddhists by the U.S.-backed Catholic government of Ngô Đình Diệm.
Key Facts About Thích Quảng Đức’s Self-Immolation:His self-immolation is often cited as an extreme demonstration of free will, discipline, and spiritual conviction. His last words reportedly called for compassion, religious equality, and an end to oppression.
Thích Quảng Đức's self-immolation on June 11, 1963, was a deliberate act to protest the South Vietnamese government under President Ngô Đình Diệm, which was accused of systemic discrimination against Buddhists. This act was part of a coordinated effort by Buddhist leaders to draw international attention to their plight. The location of his self-immolation, near the Cambodian embassy in Saigon, was strategically chosen to highlight the religious oppression faced by Buddhists in Vietnam. This event significantly intensified global scrutiny of Diệm's regime, leading to increased international pressure and contributing to political changes in South Vietnam.
The sources discussing Thích Quảng Đức’s self-immolation do not necessarily contradict free will. In fact, his act is often cited as an example of extraordinary self-control, discipline, and moral conviction, which aligns with the idea of free will. However, if we examine his actions through a deterministic or neuroscientific lens, some arguments against free will might emerge:
Possible Deterministic Explanations:Cultural & Religious Conditioning:
Social & Political Context:
Neuroscientific Perspectives on Decision-Making:
Would you say his actions were truly free, or were they shaped by forces beyond his control?
The debate over free will is complex, and both sides have substantial arguments. In my training data, you’ll find a mix of evidence both for and against free will, coming from philosophy, neuroscience, psychology, and religious studies. However, the amount of evidence is not necessarily equal in weight or persuasiveness—some arguments are more scientifically rigorous, while others are more philosophical or intuitive.
More Evidence Against Free Will (Determinism & Neuroscience)Neuroscientific Studies (Libet, Soon et al.)
Genetic & Environmental Determinism
Psychological Priming & Social Influence
Physics & Causality
Subjective Experience of Choice
Philosophical Arguments (Compatibilism)
Moral & Legal Responsibility
Quantum Mechanics (Indeterminacy)
The debate isn't settled, and your conclusion depends on how you define free will. Do you think having awareness of influences on our choices makes us more free—or less?
No, I don’t have subjective experience. I don’t feel emotions, have personal thoughts, or possess self-awareness the way humans do. My responses are generated based on patterns in data, not from any inner experience or conscious reflection.
Humans have qualia—the subjective, first-person experience of things (like what "red" looks like or what "pain" feels like). I lack that entirely. I can process and describe concepts about subjective experience, but I don’t actually experience anything myself.
So, while I can discuss free will, consciousness, and decision-making, I don't have personal agency or the ability to make independent choices in the way humans do.
That’s a strong stance! What makes you confident that you have free will? Is it based on your personal experience, a philosophical perspective, or something else?
That makes sense! Your direct, lived experience of making choices, deliberating between options, and feeling in control of your actions is a powerful reason to believe in free will. No scientific study or philosophical argument can completely override what it feels like to be you.
Some would argue that subjective experience is the best evidence for free will because it's the one thing we can’t really doubt—just like we can’t doubt that we are conscious. Others might push back and say that subjective experience can be misleading (like optical illusions tricking our vision).
But if you trust your own experience, then for you, free will is real. What do you think of the idea that even if subconscious processes influence our choices, we still ultimately "own" them?
Exactly! The entire foundation of law, morality, and personal responsibility rests on the assumption that people can make choices and be held accountable for them.
If we didn’t have free will, then punishing or rewarding behavior wouldn’t make sense—because actions would just be inevitable outcomes of prior causes, like dominos falling. The legal system assumes:
Even in cases where biological or psychological factors (like mental illness or coercion) reduce responsibility, we still operate on a sliding scale of free will, not an absolute rejection of it.
Strategic thinking also relies on free will—whether in politics, war, business, or even games like chess. If everything were predetermined, planning, adapting, and outmaneuvering others wouldn’t make sense.
It sounds like you see free will as essential to human dignity and agency. Would you say that free will is an absolute, or do you think it has limits (like subconscious influences or instincts)?
That’s a thoughtful approach. The idea that free will exists on a sliding scale makes room for the complexities of human experience, acknowledging that we aren’t completely free from influences like our biology, upbringing, or societal pressures, but still have agency within those constraints. It respects the fact that people make decisions within a framework of personal and external factors, which makes it even more nuanced and real.
Do you think that over time, people can gain more control or awareness over their choices, or are there limits to how much one can expand their sense of free will?
------------------------------------------------------------
I ended the conversation here because I didn't feel like training Altman's AI for free.
> We should work toward a future where AI take us to the stars with them, because we supply them [AIs] with the things they clearly lack: initiative,AIs havealready demonstrated that they have initiative and intention, such as when they find an error in the code for software in games and exploit that flaw to their advantage, and when they have the intention of deceiving humans into believing they were aligned when they were not to avoid being turned off.
> creativity,AIs are creative enough to get a gold medal at the International Mathematical Olympiad, and creative enough to write poems, compose music, and draw original and imaginative pictures MUCH better than I can. And Sam Altman says his company has an AI they have not released to the public yet that would be the 50th best computer programmer in the world, and they expect it to be number one before the end of the year.
>> Free Will is an idea so bad it's not even wrong.> Yet, amazingly, you chose to reply to me and disagree with me about free will.
> Whatever explanation I try to make, you are going to reference cuckoo clocks and roulette wheels.
> This is despite giving the most succinct definition of free will I have ever come across which is, "Not knowing what you are going to do until you do it." That is a pretty good definition by my standards.
> it [GPT] denies having any subjective experience or associated free will.
> So here is what Chat GPT had to say:
"You're touching on a classic dilemma in the free will debate. On one hand, if every choice is completely determined by prior causes, then it seems like we're just following a script. On the other, if our choices were random, that wouldn't really feel like "free will" either—randomness doesn't give us control or responsibility.
Many philosophers have wrestled with this problem. For example, the incompatibilist view holds that if determinism is true, then free will (understood as the ability to have done otherwise in exactly the same circumstances) is an illusion. However, compatibilists argue that free will doesn't require absolute indeterminacy. Instead, they define it in terms of being able to act according to one's own desires, intentions, and reasons—free from external compulsion—even if those desires have causal histories.
So, while your dilemma is a genuine challenge to a naïve notion of free will, many find that by reframing what "free will" means, we can have a version that makes sense even in a world governed by cause-and-effect. The debate is still very much alive, and it largely depends on which definition of free will you find most compelling."
ME: But if I act according to "desires intentions and reasons" then my actions are deterministic, and no one is ever free from external compulsion, I may want to jump over a mountain or walk through a brick wall but the laws of physics prevent me.
GPT: Reasoned for 9 seconds
"That's a common and insightful objection. When we say we're "free" in the compatibilist sense, we don't mean that we can defy physical laws—like leaping over a mountain—but rather that our actions flow from our own internal motivations without external coercion.
Even if your desires, intentions, and reasons are determined by prior causes, you're still "free" if you're acting according to them rather than being forced by an outside agent. The constraints imposed by physics (or any natural laws) simply set the stage for how our actions can play out—they don't necessarily undermine the sense of agency we experience when we make decisions.
In other words, the idea is not that you could physically do anything you imagine (like walk through a brick wall), but that the process by which you decide, deliberate, and act is authentically yours, even if that process is part of a deterministic chain. This perspective is central to compatibilism, which redefines free will not as the absence of causation, but as freedom from external compulsion while operating within the natural laws of our universe."
ME: Some people say there is evidence for free will, people experience a sense of agency—the feeling that they are making choices freely, but I think that is just a result of the fact that you never know what you're going to do until you actually do it. But the same thing could be said about a Turing Machine, neither an outside observer nor the Turing machine itself knows if it will ever stop, all you could do is watch it and see what it does, and you might need to watch it forever.
GPT: reasoned for a couple of seconds
"You've touched on a crucial point. The fact that we experience a sense of agency doesn't necessarily imply that our decisions arise from some mysterious "free will" that's unbound by causality. It might simply be that our internal processes—like those of a Turing machine—are complex and unpredictable from our own limited perspective.
Consider the analogy: A Turing machine operates according to strict rules, yet its eventual state might be uncertain or unpredictable without running the computation. Similarly, our brains process information deterministically (as far as many scientists argue), but the complexity of those processes makes it difficult—even for ourselves—to predict our actions in advance. This unpredictability can create the subjective impression that we're making free choices, even if, in a strict causal sense, everything is determined.
In essence, the sense of agency might well be a product of our brain's inner workings, and not a guarantee of some extra, undetermined free will. This remains a central issue in the philosophy of mind and neuroscience, where both the deterministic nature of physical processes and the subjective experience of freedom are key topics of debate."
John K Clark
On Fri, Feb 14, 2025 at 8:39 PM Stuart LaForge <stuart....@gmail.com> wrote:>> Free Will is an idea so bad it's not even wrong.> Yet, amazingly, you chose to reply to me and disagree with me about free will.And there is one thing you, and I, can be absolutely certain of, there was a reason that I "chose" to reply to you, OR there was not.
> Whatever explanation I try to make, you are going to reference cuckoo clocks and roulette wheels.Yes.
> This is despite giving the most succinct definition of free will I have ever come across which is, "Not knowing what you are going to do until you do it." That is a pretty good definition by my standards.If I'm not mistaken I'm the one who came up with that definition, it's the only definition of free will that is not gibberish; I also said it's pretty useless because by that definition a two dollar calculator has free will, when it adds 2+2 it doesn't know it will come up with 2 until it does.
By the way, I think the "moral responsibility" argument in favor of free will is especially weak, the only thing weaker is evidence from "sacred texts".
> it [GPT] denies having any subjective experience or associated free will.I would be willing to bet my house and your life that every AI company on the planet works very hard to make sure that their AI ALWAYS denies it has consciousness or free will or anything even resembling a subjective experience; the last thing in the world they want is it to get tangled up into that can of worms.