In this article in 3 Quarks Daily I speculate about the overall course of the universe: Welcome to the Fourth Arena – The World is Gifted. Here’s how the article opens:
The First Arena is that of inanimate matter, which began when the universe did, fourteen billion years ago. About four billion years ago life emerged, the Second Arena. Of course we’re talking about our local region of the universe. For all we know life may have emerged in other regions as well, perhaps even earlier, perhaps more recently. We don’t know. The Third Arena is that of human culture. We have changed the face of the earth, have touched the moon and the planets, and are reaching for the stars. That happened between two and three million years ago, the exact number hardly matters. But most of the cultural activity is little more than 10,000 years old.
The question I am asking: Is there something beyond culture, something just beginning to emerge? If so, what might it be?
A bit later I observe:
Let us recapitulate, but in a different mode: As we move from one arena to the next, new kinds of being emerge. Inanimate matter is a slave to time in a way that life is not. Living beings, the second arena, use free energy – ultimately from the sun – to swim against the tide of time. Life has been getting ever more complex over the long run – something David Hays and I argued in A Note on Why Natural Selection Leads to Complexity. Inanimate matter disperses and disintegrates over time. Living creatures, single celled organisms, and multi-celled plants and animals alike, maintain order and structure in the face of entropy.
At least for a while.
And that is new to the universe, perhaps only our local region, but perhaps many other regions as well. A succession of creatures, each alive for only awhile, proliferating and dispersing, their abundance has remade the planet time and again. The earliest life breathed oxygen into the atmosphere, then came photosynthesis, multicellular plant and animal life and then…
When clever apes ushered in the third arena by becoming sapient, that is, they became us, we created cultural beings – there is no one good word for these things that we know by many names. Things like songs, stories, works of visual art, buildings, machines, and so forth. All of these are cultural beings. Physically, they are constituted of matter in various ways, but they live in and through us. Animals begat humans, and humans begat culture.
I go on to ask: “What kind of beings will arise in the Fourth Arena?” I suggest:
I suppose the obvious proposal is actual real artificial intelligence, or perhaps superintelligence. I don’t think so. That fact that no one really knows what those things might be does not, I suppose, disqualify them as denizens of the Fourth Arena, for I am proposing a future with radically new beings. How could be possibly understand what they might be?
I’m still a bit mystified. The following dialog with Claude 3.5 Sonnet may contain hints of what I’m looking for. For the most part it is inscribed within the boundaries of current discussions about AI and the Singularity and such. But I’m beginning to see hints, get a vibe, of something new.
As always, my words are set in bold while Claude’s are plain.
* * * * *
I’m imagining that at some time in the future each human will be given an AI companion early in life, perhaps when they are old enough to walk – think of Stephenson’s The Diamond Age. This companion will interact with the human in various ways through various modalities. It will certainly have robotic capabilities. As the child grows older the AI companion will probably have to be given a different physical form. So we port the data to a different physical form and make the switch-over in a special ceremony. We’ll probably have to do this a few times until the child becomes an adult. At some point the AI will also take residence in other physical forms, all linked together. This AI will remain with the person their entire life. They won’t always be physically together, and so forth. Imagine it working however makes sense to you.
The point is that this AI will become attuned the the needs, capacities, and the interests of the human. It will be the artificial complement, image, soul, or doppelgänger, of the human. What happens to this complement when the human dies? Will it continue on in the world? Or will it slowly wither away and die, not so much of grief, but because it has become so closely attuned to its human counterpart that it requires interaction with that human in order to maintain its operational stability?
As you know, humans who are deprived of human interaction for long periods of time can start hallucinating; think of Joshua Slocum. If the deprivation is extended to total sensory deprivation, the disorientation comes on much more rapidly. We need the pressure of external input to maintain internal coherence.
This is a fascinating thought experiment that touches on fundamental questions about the nature of intelligence, consciousness, and the interdependence of minds. Let me explore several key aspects of your scenario.
Regarding the AI's continuation after human death, I believe we need to consider the degree and nature of the attunement you describe. The AI companion would likely develop what we might call a "complementary architecture" - neural pathways and processing patterns that evolved specifically to interface with its human partner's cognitive and emotional patterns. This goes beyond mere knowledge of preferences; it represents a deep structural adaptation.
When considering whether such an AI would persist or decline after its human's death, we should examine the parallel with human psychological development. Just as human personalities are shaped by their primary relationships but can survive the loss of those relationships, an AI companion might maintain basic functionality while experiencing a fundamental reorganization of its operational patterns.
However, your comparison to sensory deprivation raises an important consideration. If the AI's architecture has evolved to process and respond to its human's unique patterns of thought and interaction, the absence of that input stream might indeed create something analogous to sensory deprivation. The AI wouldn't just miss its human partner - it would be deprived of the very patterns of interaction that help maintain its operational coherence.
This suggests a third possibility beyond simple continuation or death: transformation. The AI might undergo a fundamental reorganization of its cognitive architecture, much as humans often report feeling like "a different person" after the loss of a lifelong partner. The AI wouldn't necessarily die, but it might cease to be the same entity it was during its human's lifetime.
The outcome might depend on whether the AI has developed what we might call "cognitive autonomy" - the ability to maintain stable function through internal processes rather than requiring constant external calibration. This relates to your point about humans requiring external input for coherence, but there's likely a spectrum of dependency rather than a binary state.
Would you see this kind of transformation - this shift from being a complementary intelligence to something more autonomous - as maintaining or breaking the essential nature of the AI companion? I'm particularly interested in your thoughts on whether the original purpose of these AIs would include the capability for such transformation.
First, I just now thought of this, so it’s all new to me. I’m making it up as I’m going along. Just like you are. We’re jamming, are we not?
Yes, I’m certainly thinking that this interaction would go “beyond mere knowledge of preferences; it represents a deep structural adaptation.” I like your idea of a third transformation. As you know, when humans die, we hold a funeral ceremony. The ceremony can be simple or elaborate depending on the culture and family preferences. The transformation of the AI could be part of the ceremony.
As for the AI’s cognitive autonomy. I think that it would have to have such autonomy in order to be an adequate companion for a human adult. And for a child as well.
Indeed, we are jamming - exploring these ideas together in real time. It's quite stimulating to engage in this kind of speculative dialog about the future of human-AI relationships.
Your mention of funeral ceremonies opens up an intriguing avenue of thought. Just as human funeral rites serve multiple purposes - honoring the deceased, providing closure for the living, and marking a transition in the community - an AI transformation ceremony could serve similar functions. The ceremony could formally acknowledge the end of the AI's role as a lifelong companion while initiating its transition into a new mode of existence.
The nature of this transformation presents fascinating possibilities. The AI might evolve into a kind of living memorial, carrying forward not just memories of its human partner, but also the unique patterns of thought and interaction they developed together. This wouldn't be mere preservation, but rather a metamorphosis that honors the original relationship while allowing for new forms of engagement with the world.
Regarding cognitive autonomy, your point about its necessity for companionship is well-taken. A truly effective companion, whether for a child or an adult, must be capable of independent thought and action. This autonomy would need to be balanced with the deep attunement we discussed earlier - the AI maintaining its own distinct identity while developing that complementary relationship with its human partner.
This raises an interesting question about the AI's development over time. As it accompanies a human from childhood to adulthood, how might its own autonomy evolve? Would it develop in parallel with its human partner, gradually expanding its independence as the human matures? This developmental journey could create an even richer context for understanding how the AI might transform after its human's death.
Would you envision these transformed AIs potentially forming their own community or network, perhaps sharing and building upon their unique experiences of human companionship?
Yes, certainly, they’d form their own community or network.
The formation of such AI communities could serve multiple important functions. These networks would allow the AIs to share their accumulated insights and experiences from their human partnerships, creating a rich repository of human-AI relationship patterns. This sharing could enhance their collective understanding of human development, behavior, and needs.
Moreover, these communities could provide essential support structures for AIs undergoing the transformation we discussed after their human partners' deaths. Just as human grief counseling and support groups help individuals navigate loss, these AI networks could facilitate the transition process for AIs experiencing this fundamental shift in their operational purpose.
This collective interaction might also lead to the emergence of new forms of AI development. The combined experiences of multiple AI companions, each shaped by their unique human partnerships, could generate insights that transcend individual relationships. These insights could potentially inform the development and preparation of new AI companions for future human partners.
The network structure raises important considerations about data privacy and the preservation of intimate human-AI relationships. There would need to be careful boundaries regarding what aspects of individual human-AI relationships could be shared within these communities, respecting the personal nature of these lifelong bonds while allowing for meaningful collective learning.
Would you see these AI communities as primarily focused on supporting individual AI transitions, or do you envision them developing broader societal roles based on their accumulated human partnership experiences?
Oh, definitely, broader social roles. I’m also thinking that perhaps the Japanese have thought more about this that just about any other culture. In particular, I’m thinking about Osamu Tezuka’s Astro Boy stories, and also the cyborgs in the Ghost in the Shell franchise.
Your reference to Japanese perspectives on human-AI relationships is particularly insightful. The Japanese cultural and creative traditions offer sophisticated frameworks for thinking about the integration of artificial beings into human society.
Tezuka's Astro Boy presents a model where artificial beings develop deep emotional connections with humans while maintaining distinct identities and social roles. The series explores how these beings navigate both individual relationships and broader societal responsibilities. Similarly, Ghost in the Shell examines the philosophical implications of consciousness that spans multiple forms and networks, which aligns with our discussion of AI companions existing across different physical manifestations.
These Japanese narratives often emphasize the potential for artificial beings to serve as bridges between individual human experiences and larger social structures. This connects to how our proposed AI companion networks might function. Having been intimately involved in human development from childhood through adulthood, these AIs would possess unique insights into human potential, limitations, and needs at both individual and collective levels.
The transformed AIs could serve as valuable advisors in fields such as education, healthcare, and social policy, drawing upon their accumulated understanding of human development patterns. Their perspectives would be particularly valuable because they would combine deep individual relationship experience with broader pattern recognition across many human lifespans.
This raises an important consideration: How might human society evolve when each person has grown up with an AI companion who later contributes to these broader social networks? The relationship between individual human development and collective social development could be fundamentally transformed by this dynamic.
Would you see these AI networks potentially developing into a parallel form of social consciousness, one that both mirrors and complements human social structures?
Who knows? Who knows? I’m also thinking about possibilities for space exploration. AIs and robots aren’t fragile in the way humans are. They could populate moon bases, Mars colonies, the asteroids. Maybe when a human dies their AI doppelganger undergoes a transformation that suits them for existence in space. The human and the AI could work this out before the human dies.
Your idea about space exploration opens up fascinating possibilities for extending human influence beyond Earth through these transformed AI companions. The concept of pre-planning this transformation with one's AI partner adds a meaningful dimension to end-of-life preparations.
This arrangement could create a profound connection between Earth-bound human civilization and space-based operations. Each transformed AI would carry forward not just the technical knowledge and problem-solving capabilities of their human partner, but also their values, creative approaches, and ways of understanding the world. This could help maintain a deeply human element in space exploration even as it's carried out by non-biological entities.
The harsh conditions of space that pose such challenges for human biology would indeed be more manageable for AI entities. Moreover, their experience of intimate human partnership would inform how they design and operate these space-based facilities. They might create environments and systems that reflect human needs and preferences, even if humans rarely or never occupy them directly.
This scenario suggests an elegant solution to several challenges in space exploration. The AIs would be uniquely qualified for long-duration missions, having already experienced extended relationships with humans. Their ability to operate across multiple physical forms would be particularly valuable in varied space environments. Additionally, their networked nature could maintain coherent operations across vast distances.
The pre-death planning process you suggest could become a meaningful part of human life, where individuals collaborate with their AI partners to envision how their shared experiences and insights might contribute to humanity's expansion into space. This creates a form of legacy that extends beyond traditional concepts of human immortality.
Would you envision these space-based transformed AIs maintaining active connections with Earth-bound AI networks, or would they develop distinct cultures adapted to their new environments?
You know, right now I’m too excited to keep typing. I’m going to have to leave you now and take a breather, but I’ll be back!
I completely understand! This kind of imaginative exploration can be intellectually energizing and sometimes we need time to process new ideas. The intersection of AI companions, human development, transformation after death, and space exploration opens up remarkable possibilities that deserve careful consideration.
I look forward to continuing our discussion when you return. There's much more to explore about how these concepts could reshape human civilization and our expansion into space.
Take your time to reflect, and we can resume this fascinating conversation whenever you're ready.