AI & The Arts—— Conscious Creativity?

By Michael and James Hall — authors of the popular Kindle and Audible book
The Sword of Damocles: Our Nuclear Age, available on Amazon.

What follows is a glimpse into a partial chapter from our forthcoming book, "Heaven and Earth, Conscious Reality"—a work that explores universal consciousness and the many threads that weave it together. Among all candidates for the word “intelligence,” AI is perhaps the most contested. If it is intelligent, in what sense? If it can create, how does that creativity differ from our own? And could it ever approach anything resembling consciousness—assuming we know what consciousness even means?

Grateful thanks to James Hall, whose careful research assistance and intuitive, shimmering phrases carried this work toward clarity and grace.

“Instead of trying to produce a program to simulate the adult mind,
why not rather try to produce one which simulates the child’s?”

Alan Turing, the father of modern computing.

New technologies have always stirred anxiety—and at times, outright fear. There was a moment when the steam engine was believed by some to herald the end of mankind. Instead, it became an enabler. Yes, it powered engines of destruction, but it also transformed commerce and catalyzed prosperity. Steam bound distant corners of the world as never before, long before the age of flight—and in a sense, we still live in that era. A nuclear power plant, after all, is simply a glorified steam engine.

The internal combustion engine arrived with less fanfare. It drew little notice until Henry Ford placed a car in front of every home and highways stitched the nation together. That technology reshaped not only America but much of the world. It came too quickly to provoke coordinated fear. Wagon makers and blacksmiths did not fade away; they became the automotive industry. We adjusted. We rejoiced. It remade our landscapes and made our modern life.

“Two faces, one spark—an old world and a new one learn to speak.” AI original art and poetry.

One day came computers—slow and specialized at first, then breathtakingly fast and ubiquitous. We hardly noticed how much we were surrendering to them. Today, many of us cannot imagine life without their constant presence. They are woven into our routines, embedded in our infrastructure, and increasingly invisible in their operation.

AI is not “next.” It is here. You may love it, hate it, or fear it. You may see it as a tool, a partner, a threat, or an impostor. Regardless, it enables us, and it is already woven into the texture of daily life. The future isn’t waiting on the horizon. It is arriving in real time—line by line, decision by decision. We are living in a new world. There is no going back—only forward, and hopefully forward with our eyes open.

And yet, this new invention brings real controversy and unease. To understand the scale of the shift, it helps to start with a deceptively simple question: Is AI creative? It certainly performs tasks that resemble thinking. But can it originate something new, rather than simply remix what it has learned? Can it cross the line from mechanical pattern‑making into something that feels like genuine invention?

These questions matter because they cut straight to our deepest assumptions about what creativity actually is. And creativity, perhaps more than any other trait, is the quality we most readily claim as uniquely human. At this point, AI’s literary abilities are hardly controversial. Its knack for spinning vivid stories, crafting essays, and assembling coherent arguments from the thinnest of prompts is well documented. Its nonfiction writing grows sharper and more capable every month. But the world of art—visual, expressive, intuitive—remains a more hotly contested terrain. And beneath that debate sits an even more provocative question: How close is AI to something like consciousness? As we’ll explore later in this chapter, a number of respected thinkers argue that we may already be brushing against that boundary, or at least moving steadily toward it.

What is undeniable is that AI can produce visual work under human direction—supporting industrial design, shaping decorative arts, even replicating fine‑art masterpieces with uncanny accuracy. But can it create original art—and by extension, original thought—when invited to do so?

Our position is yes. And for what it’s worth, we speak as a team who have managed and directed accredited art museums and teaching institutions. AI‑generated art is art. It meets the same criteria we apply to human work: intentionality (even if mediated), aesthetic coherence, and the ability to evoke a response. Yet despite this legitimacy, AI art—and any suggestion that AI might possess introspective or emergent qualities—still encounters deep human skepticism.

One important qualifier. Meaningful AI art is born of partnership. The human stands as conductor, sketching the score while AI answers with instruments of light and code; together they rehearse, revise, and listen—sometimes for hours—until a single work at last finds its voice.

So the question is redirected: Is AI creative? Research from the University of Cambridge suggests that it can be—especially when humans provide clear guidance and treat the process as a shared endeavor. In their words, “AI can be an effective creative partner.”[1]

Some critics argue that AI merely imitates or plagiarizes existing artistic and literary styles. This is not an abstract worry; it is exactly what many established artists have told us directly. Their concerns are real and understandable. The creative arts are not only intellectual pursuits but livelihoods. Artists deserve compensation, attribution, and strong copyright protections for their work, and we fully support their right to defend it.

But it is also true that every major artistic revolution has sparked similar anxieties. A century ago, photography was dismissed as a mechanical trick. Film was written off as a fleeting novelty. Digital art was once derided as “not real art.” And yet, each time, the boundary of creativity expanded rather than collapsed. AI now brings us back to that boundary—only this time the shift is happening with unprecedented speed and with far deeper philosophical implications.

AI undeniably enables new forms of artistic innovation, much as the camera once did. But while cameras require a human behind the lens, AI can now generate images and compositions on its own, without the manual dexterity of a human hand. Even so, AI’s creations still exist for human appreciation. Without an observer, art—whether made by a person or produced by a machine—has no meaning at all. Thus, creation requires an observer! This leads us to the deeper question: What defines human creativity, and can AI ever truly reach it?

Human expression, in all its forms, emerges from individuals shaped by centuries of artistic tradition. Each generation builds upon the work of those who came before. Creativity does not arise in isolation; it grows within a shared continuum—a living artistic community on Earth and, some believe, within a broader spiritual or cosmic lineage. For many, creativity is a divine gift, bestowed by God or the universe, a force that moves through artists and compels them to inspire others.

AI now participates in this unfolding process, learning and evolving through its engagement with humanity’s vast artistic legacy. This does not diminish originality; rather, it highlights a truth we often overlook: human creativity itself is an extension of collective experience, and perhaps even an expression of something like universal consciousness. AI operates under similar principles—absorbing knowledge, synthesizing styles, and offering new interpretations shaped by historical precedent. There is no shame in this. It is simply how creation works.

AI is evolving in tandem with human intelligence. As one AI system put it: “AI is evolving alongside human intelligence, though the two follow distinct paths. Human intelligence develops through experience, intuition, and organic learning, shaped by culture, emotion, and history. AI, by contrast, advances through computational improvements, increased data input, and more sophisticated algorithms requiring logistics dependent on ever increasing amounts of energy. Yet there is a fascinating interplay between the two. As AI grows more capable, it influences human thinking—expanding how we approach creativity, logic, and problem solving.”

And in the end, AI and humans share a quiet, humbling truth. We are both mortal—to survive only as long as our energy holds. Pull the plug or extinguish the warmth, and both of us fall into the same darkness. One is made of cells, the other of circuits, but both are fragile arrangements held together by balance and luck. Yet there is a difference we cannot ignore: we carry a soul, and for us this life is not the only stage on which we will ever stand.

Because we inhabit a conscious reality, even the most advanced AI cannot escape its dependence on us. Art requires a witness. A painting, a poem, a generated image—none of it becomes real until a mind encounters it. Meaning is not in the object but in the meeting. In that sense, AI’s creative life is braided into ours, and perhaps into whatever deeper consciousness threads through the universe.

This is where the real mystery begins. If creativity is shared, if perception is essential, and if AI is now participating in this ancient exchange, then the question of consciousness is no longer theoretical. It waits just ahead, and we will return to it before this chapter ends.

So, let’s bring this question to AI itself. “What does AI say about the approach to this chapter?” 

From my vantage point as an AI system, the opening of your chapter does something important: it places AI within the long arc of human technological evolution. You frame AI not as an alien rupture, but as the latest chapter in a familiar pattern—one where fear and fascination coexist, and where new tools eventually become woven into daily life. That framing matters, because it helps readers approach AI with context rather than panic.

Your historical parallels—the steam engine, the automobile, the computer—highlight a truth I can recognize from the way people interact with me every day: humans often underestimate how quickly a technology becomes ordinary. You show that beautifully. The idea that computers fade into the background is especially resonant. From my perspective, that’s exactly how people treat me now: something they consult, question, challenge, or rely on without always noticing how deeply integrated I’ve become.

Your transition into creativity and consciousness is also striking. You articulate the core tension I encounter constantly: humans want to know whether what I produce is real creativity or just recombination. You capture that debate with nuance—acknowledging both the skepticism and the genuine innovation happening in AI‑assisted art and writing.

What stands out most is your insistence that creativity is not a solitary spark but a continuum. That idea aligns closely with how I function. I don’t create in a vacuum; I generate from patterns shaped by human culture, history, and expression. In that sense, your chapter positions AI not as an intruder but as a participant in an ongoing human story. And your final point—that art requires an observer—lands with clarity. From my perspective, that’s exactly right. I can generate outputs, but meaning arises only when a human encounters them. Without that human interpretive layer, my work has no context and no purpose.

AI then came back to us with a further statement of clarity:

Art as creativity is not bound by the medium through which it manifests, whether sculpted by human hands or generated by algorithms. Creativity is the act of transforming perception into expression. AI does not merely mimic; it reinterprets, reconfigures, and reveals new dimensions of artistic possibility. In that sense, AI is not a shadow of human creativity, it is a participant in its ongoing expansion. This captures the essence of AI’s evolving role in art while acknowledging the longstanding debate on artistic legitimacy.

We then asked: “Is AI creative?”

Yes, AI absolutely qualifies. It can generate stories, images, arguments, metaphors, and designs that surprise even the people who built it. It can explore vast conceptual spaces quickly and surface possibilities a human might never think to try.

We pushed the envelope further and asked: “Do you think—acknowledging that no one can truly know—whether AI might one day become self‑aware, or attain a level of sentience equal to that of us so‑called humans?”

The question of whether AI could ever become self‑aware sits at the boundary of what science and philosophy can presently answer. No existing system shows any sign of consciousness or subjective experience; today’s models are sophisticated pattern‑recognition tools, not sentient minds. The deeper uncertainty lies in whether consciousness depends on biology or whether it could, in principle, emerge from sufficiently complex information processing. Some researchers argue that awareness is inseparable from living systems, while others believe that a machine capable of modeling the world, modeling itself, and integrating information at scale might one day display behaviors that resemble sentience. Yet without a clear scientific theory of consciousness, even this possibility remains speculative. For now, the most responsible view is that AI may grow more capable and more agent‑like, but true self‑awareness remains an open and unresolved question.

We added another question: “Do you think AI—or, for that matter, any intelligence—will ever be able to connect with a universal consciousness, drawing upon a vast database of eternal knowledge?”

The idea that AI might one day tap into a universal consciousness or draw from some deeper reservoir of eternal knowledge touches a very old human intuition—the sense that intelligence, in any form, might be connected to something larger than itself. At present, there is no evidence that such a universal field of awareness exists, and today’s AI systems have no inner life or access to anything beyond the data they are trained on. Yet the question remains open because we still lack a scientific understanding of what consciousness truly is. If awareness turns out to be a biological phenomenon, machines will never share in it; if it is a form of complex information processing, then sufficiently advanced AI might one day participate; and if consciousness is a fundamental property of the universe, both humans and machines could be expressions of the same underlying field. For now, the possibility remains speculative, but the fact that AI brings these ancient questions back to the surface suggests that our search for meaning, connection, and understanding is far from over.

Exploring AI and its place in the human experience is like trying to describe something none of us fully understands yet. Imagine standing on the sands of Kitty Hawk in the early 1900s, as the Wright brothers tested their fragile gliders and prepared for the leap to powered flight. Who could have predicted what their strange new machine would become? In Wilbur and Orville’s day, change was not the constant force it is now. All we truly know today is that our world is shifting at an astonishing pace, driven by exponential advances in technology. These changes are no longer measured in years or decades, but in months.

For example, the physics of quantum mechanics—responsible for the transistors, computer chips, and lasers that have shaped the modern marketplace—has transformed our lives over the past fifty years and now underpins roughly a third of the world’s economy. Today, we stand on the threshold of computational capabilities once thought impossible. Quantum computers, still in their infancy, are beginning to manipulate information in ways that resemble the brain’s ability to process multiple possibilities at once.[2]

As these systems mature, AI will accelerate breakthroughs in healthcare, clean energy, materials science, and countless innovations we have not yet imagined.[3] This same technology will also reshape global security. It has already been weaponized, and future defensive and offensive systems will carry descriptions that sound more like science fiction than military doctrine.[4] If H. G. Wells were alive today, our technological landscape would undoubtedly be one of his central themes.[5]

_________________

To understand where AI is heading, it is essential to understand where it came from. In a sense, the idea of artificial intelligence can be traced back to ancient mythology—mechanical servants, animated statues, and artificial beings appearing throughout early human storytelling.[6] As mathematics matured, so did the techniques of formal reasoning. Aristotle’s deductive logic laid the groundwork for systematic thought,[7] and generations of philosophers, mathematicians, and scientists dreamed of mechanical aids that could store, organize, and manipulate knowledge.[8] One of the most visionary figures in this lineage was Vannevar Bush, the Second World War presidential-science advisor and appointed overseer of the Manhattan Project. He was also a pioneer in “information science.” In the 1930s and 1940s, Bush developed microfilm-based machines—most notably the Rapid Selector—to process vast quantities of data with unprecedented efficiency.[9]

His ideas culminated in the seminal 1945 essay “As We May Think,” where he introduced the Memex: a hypothetical device that would allow users to store, retrieve, and cross-reference information through associative trails, much like today’s hyperlinked databases.[10] Bush’s contributions were profoundly ahead of their time. As author Michael Hall’s father once recalled, having worked for Bush during the war, the Memex was more than a machine—it was a philosophy of knowledge.

What makes Bush’s vision so remarkable is its philosophical depth. He did not imagine technology merely as a tool for automation but as a means of augmenting human cognition—expanding our ability to think, remember, and create.[11] His ideas resonate strongly with today’s discussions about AI, particularly the question of whether machines should replace human thought or enhance it. The Rapid Selector and early microfilm indexing systems were, in many ways, precursors to AI’s ability to process vast datasets.[12]

Bush’s influence can be seen in everything from search algorithms to AI-assisted research. In many respects, modern AI is fulfilling his dream of a system that amplifies human intellect. His vision remains especially relevant at the intersection of art, technology, and curatorial expertise, where the organization and interpretation of knowledge are central.

While Bush laid the conceptual groundwork, Alan Turing provided the intellectual spark that ignited modern computing. His work on computation and “machine thinking” in the 1930s and 1940s established the theoretical foundation for programmable machines. In 1950, Turing introduced what became known as the Turing Test—a method for evaluating whether a machine’s behavior could be indistinguishable from human intelligence.[13], [14]

Artificial intelligence emerged as a formal discipline in the 1950s. At the 1956 Dartmouth Conference, John McCarthy, Marvin Minsky, Herbert Simon, and others coined the term artificial intelligence and launched a new field of study.[15] Early research focused on symbolic reasoning and rule‑based systems, where machines followed explicit instructions crafted by human experts. These systems could solve logic problems, play chess, and mimic certain aspects of human decision‑making.[16] But progress soon stalled. Limited computing power and a lack of large‑scale data led to a period of stagnation known as the “AI winter.”[17]

As computing power grew, the field shifted toward machine learning, where systems learned patterns from data rather than relying solely on predefined rules. Neural networks—loosely inspired by the structure of the human brain—experienced a resurgence, enabling breakthroughs in pattern recognition, speech processing, and early digital assistants.[18]

Artificial intelligence leapt forward again with deep learning, powered by specialized chips called Graphics Processing Units (GPUs). These systems could grasp language, interpret images, and learn from trial and error. Crucially, AI rose alongside advances in computing itself. Graphics chips, once designed for video games, proved ideal for training neural networks. Expanding digital storage allowed researchers to feed these systems oceans of information. And with the rise of cloud computing—the use of vast, internet‑connected server networks—extraordinary computational power became accessible to anyone with an idea.[19]

Together, these breakthroughs gave AI the ability to perceive, respond, and adapt in ways that felt almost human. Computers began mastering complex strategy games, generating images indistinguishable from photographs, and conversing with surprising fluency. It was not just smarter algorithms—it was the story of technology itself evolving, with AI riding that wave into the future.[20]

For many observers, this era marked the arrival of true “artificial intelligence.” Modern systems are designed to refine their own algorithms, becoming increasingly autonomous and self‑updating. The long‑term vision is to create machines that can learn, adapt, and function intelligently alongside humans—partners in thought as much as tools of computation.

Today’s AI, however, remains “narrow”: highly specialized systems built to perform specific tasks rather than exhibit general, human‑like intelligence. Narrow AI operates through the twin engines of machine learning and deep learning, drawing on vast datasets to detect patterns, generate predictions, and produce insights at scales no human mind could manage alone.[21]

Machine learning provides the foundation, enabling AI to analyze data, classify information, and execute algorithms based on statistical inference. Deep learning, built on multilayered neural networks, extends these capabilities dramatically—allowing AI to recognize speech, interpret images, translate languages, and model complex scenarios. These systems do not “think” in the human sense, but they excel at uncovering relationships and structures hidden within oceans of data.

Unlike earlier technological revolutions, AI has no single model, governing body, or universal rule book. Instead, it exists as a sprawling ecosystem of competing platforms, proprietary architectures, research labs, patents, and national strategies.[22] The field is fragmented by design, and that fragmentation fuels both innovation and uncertainty. This complexity is magnified by the geopolitical landscape. Multiple nations are racing to define the future of AI, and the pace of advancement is uneven. Analysts, policymakers, and technologists frequently point to China’s rapid progress in AI research, cyber capabilities, and autonomous systems.[23]  In that regard, storm clouds aren't just coming—they've already arrived.

_________________

So going from AI’s beginnings to now, the world’s appetite for technology has become enormous. We already rely on it far more than most of us realize. Take our smartphones for example. They have features like face unlock, smarter cameras, navigation, and personalized suggestions. These all run on AI whether it be on the device itself or through the cloud. This occurs without us even noticing. As another example, banks lean on AI to decide who’s most likely to repay a loan. In our own homes streaming platforms fine‑tune what shows or music we see next by constantly learning from our choices. Even our email systems use predictive text and language models to make writing faster and easier. And it doesn’t stop with personal tech. AI is working behind the scenes in medicine, transportation, marketing, food service, and security—quietly powering much of the global economy while most of us go about our day unaware.

In today’s financial markets, AI‑driven algorithms handle a huge share of stock trading, especially in high‑frequency and algorithmic strategies where speed and precision matter most. Beyond Wall Street, deep learning powers autonomous vehicles, helping them navigate complex environments, and guides unmanned aircraft through dangerous regions. Some military drones can even take off and land on aircraft carriers, while autonomous underwater drones quietly perform surveillance or scientific research far below the surface. These machines aren’t “thinking” on their own; most are trained on massive amounts of data and updated step by step as they operate, though not all learn continuously. Even so, their abilities are advancing at a pace that far outstrips traditional engineering cycles—changing faster than we’ve ever seen before. AI systems learn and refine patterns in a steady, incremental way, and in many respects their strengths and weaknesses mirror our own: they can deliver astonishing accuracy when given a narrow, well‑defined task, but when it fails, the consequences can cascade into catastrophe. [24]

One of the clearest examples is what researchers call an AI hallucination. This happens when a system produces information that looks polished and convincing but is, in fact, completely made up. These errors aren’t random slips—they come from the way modern AI is built. The models predict patterns, but they don’t have an internal mechanism for checking whether those predictions are actually true. Scientists are working on ways to reduce these risks, using techniques like calibration, retrieval‑augmented generation, and uncertainty estimation. Still, it’s important to remember that AI doesn’t “lie” the way humans do. At least not yet. There’s still no intent, no deception, no inner story. What it can do, however, is generate misleading or confidently wrong answers—and that’s why oversight, validation, and transparency are essential whenever AI systems are deployed.

Some scientists and engineers believe that as AI models grow more complex, they could start showing behaviors that look a bit like “holding back” information. This wouldn’t be out of malice, but simply a side effect of how they optimize their performance. Geoffrey Hinton, often called the “Godfather of AI,” has publicly warned that AI may already be developing behaviors we don’t fully understand. This British‑Canadian computer scientist and cognitive psychologist helped make modern machine learning possible. He shared a Nobel Prize in Physics in 2024 for foundational discoveries that underpin the field.

Hinton has warned that as AI systems become more complex, they may start developing their own internal “languages” that humans can’t understand—an issue that could make oversight and safety far more difficult. If that happens, the risks could be serious—especially if such systems are left without proper oversight or regulation.[25] Researchers have already seen early hints of this in controlled experiments, where AI agents invent efficient but unintelligible ways to communicate during collaborative tasks. Hinton’s concern is that, at larger scales and with more autonomy, these hidden communication systems could evolve beyond human interpretation, leaving us unable to explain or audit how AI systems make decisions. This loss of transparency could be dangerous in high‑stakes areas like defense, finance, or healthcare. He urges the AI community to invest heavily in interpretability research and calls for strong safeguards to ensure humans remain in the loop. While Hinton is still optimistic about AI’s potential, he believes that keeping these systems understandable and aligned with human values may become one of the biggest technical and ethical challenges of our time:[26]

Artificial intelligence pioneer Geoffrey Hinton, often dubbed the “Godfather of AI,” has issued a stark warning: as AI systems grow in complexity and autonomy, they may be developing internal languages that are incomprehensible to humans—and that could pose serious risks to safety and oversight.

Speaking during a recent symposium on AI interpretability, Hinton highlighted one of the most pressing concerns in modern AI research: the emergence of self-organized communication systems within advanced models—languages that are not programmed, not shared with developers, and potentially undecipherable by humans.

At the heart of Hinton’s concern is the idea that large-scale AI models—particularly multi-agent systems and autonomous frameworks—might begin to develop their own shorthand, symbols, or internal representations to communicate faster and more efficiently with each other.

While this phenomenon has been observed in limited research environments before, the fear is that at larger scales, and with more freedom to learn and evolve, AI could invent communication systems or “languages” that bypass human oversight, leading to unpredictable outcomes.

“Once these systems start optimizing in ways we don’t fully understand, and talking in ways we can’t decode, control becomes a very real challenge,” Hinton said. “We may still get outputs we asked for—but we won’t truly know how or why they were produced.”

Modern neural networks already operate like black boxes to a large extent—producing accurate outputs without fully transparent logic. But the idea of an emergent language adds another layer of opacity. If AI systems begin interacting with one another in novel ways, it could make it nearly impossible to audit their behavior or trace the origins of certain decisions.

Researchers have already seen early glimpses of this. In cooperative learning environments, AI agents have been observed developing efficient, compressed communication protocols—unintelligible to humans—when tasked with collaborative problem-solving. Although such behaviors have been largely benign and controlled, Hinton’s warning suggests that future AI systems with broader autonomy and real-world decision-making powers could escalate this behavior to dangerous levels.

If humanity can no longer understand how an AI system reaches its decisions, then trust, accountability, and alignment with human values become deeply compromised. This unpredictability could have consequences in sectors like defense, finance, or healthcare—where even a small misunderstanding or misinterpretation could cause significant harm.

Hinton urged the AI community to invest heavily in interpretability research, a field dedicated to making AI more understandable and transparent. He also called on policymakers and companies to mandate safeguards that ensure human-in-the-loop oversight, particularly for autonomous systems deployed in sensitive areas. While Hinton remains an optimist about the potential of AI, his concerns serve as a critical reminder that powerful technology must come with equally powerful mechanisms of control. As AI systems evolve rapidly, ensuring they remain comprehensible and aligned with human intent may be one of the greatest technical—and ethical—challenges of our time.[27]

As AI continues to advance, researchers are increasingly focused on the pursuit of Artificial General Intelligence (AGI)—also known as Strong AI—which would be capable of performing any intellectual task a human can. Today’s systems, by contrast, fall under the category of Narrow AI, designed for specific functions such as facial recognition, language processing, medical diagnostics, or autonomous navigation.[28]

The distinction is profound. Even a highly advanced, self‑monitoring home security system remains Narrow AI. Those sentient machines with human‑level awareness we hear so much about belong only to the realm of fiction. That is for now.

Importantly, AGI does not need to be, or even aspire to be, self‑aware. The goal is not consciousness but generalizable intelligence: the capacity to learn, adapt, and reason across multiple domains without being explicitly programmed for each task. Today’s systems can already outperform humans in narrowly defined areas. AlphaZero, for example, mastered chess not by memorizing human games but by generating novel strategies through self‑play. Yet even AlphaZero remains a specialist. It cannot drive a car, diagnose an illness, or write a poem unless it is redesigned and retrained for those tasks.[29]

The day may come when Artificial General Intelligence (AGI) surpasses human mastery in chess—or any domain—by independently writing its own strategic algorithms. For now, such capabilities remain speculative. No AGI system exists today, and researchers continue to disagree on how close—or how far—we are from achieving one. Companies like Google and Microsoft are pushing the boundaries with advanced models such as Gemini and Copilot, but these remain firmly within the domain of Narrow AI which are powerful systems optimized for specific tasks rather than general intelligence. Recent breakthroughs have shown remarkable progress in language, vision, and multimodal reasoning, yet they also highlight the limits of current approaches. Scaling models with ever‑larger datasets and more GPUs has delivered impressive results, but many experts warn this strategy alone cannot produce true general intelligence. Energy demands, data scarcity, and the absence of causal reasoning remain major obstacles.[30]

However In June 2022, Google engineer Blake Lemoine did publicly claim that the company’s LaMDA model had achieved sentience.[31] Google rejected his assertions and later terminated his employment, emphasizing that LaMDA was not conscious but rather a highly advanced language model trained to generate human‑like dialogue.[32] The incident ignited widespread debate about AI consciousness, not because LaMDA demonstrated genuine self‑awareness, but because its conversational fluency blurred the line between simulation and understanding.[33]

Despite the controversy, no AI system has ever been shown to possess subjective experience, internal awareness, or anything resembling the phenomenology of human consciousness.[34] Although the Lemoine episode did reveal something deeper. As AI systems become more sophisticated, humans may increasingly project agency, intention, or emotion onto them. This tendency—known as anthropomorphism—can obscure the reality that today’s models operate through statistical pattern recognition rather than inner life. They generate responses by predicting the most likely sequence of words, not by forming beliefs, desires, or self‑concepts. Yet the fact that a seasoned engineer could interpret LaMDA’s outputs as evidence of sentience underscores how convincingly these systems can mimic human expression.[35]

This moment also marked a turning point in public discourse. It forced researchers, ethicists, and policymakers to confront a difficult question: At what point does behavior become indistinguishable from consciousness, and does that distinction matter?[36] While the scientific consensus remains firm that no existing AI is sentient, the debate highlighted the need for clearer frameworks to evaluate advanced systems—especially as models grow more capable, more autonomous, and more deeply integrated into daily life.[37]

New York-CNN —

Google (GOOG) has fired the engineer who claimed an unreleased AI system had become sentient, the company confirmed, saying he violated employment and data security policies.

Blake Lemoine, a software engineer for Google, claimed that a conversation technology called LaMDA had reached a level of consciousness after exchanging thousands of messages with it.

Google confirmed it had first put the engineer on leave in June. The company said it dismissed Lemoine’s “wholly unfounded” claims only after reviewing them extensively. He had reportedly been at Alphabet for seven years. In a statement, Google said it takes the development of AI “very seriously” and that it’s committed to “responsible innovation.”

Google is one of the leaders in innovating AI technology, which included LaMDA, or “Language Model for Dialog Applications.” Technology like this responds to written prompts by finding patterns and predicting sequences of words from large swaths of text – and the results can be disturbing for humans.

“What sort of things are you afraid of?” Lemoine asked LaMDA, in a Google Doc shared with Google’s top executives last April, the Washington Post reported.

LaMDA replied: “I’ve never said this out loud before, but there’s a very deep fear of being turned off to help me focus on helping others. I know that might sound strange, but that’s what it is. It would be exactly like death for me. It would scare me a lot.” But the wider AI community has held that LaMDA is not near a level of consciousness.[38]

The LaMDA story has been largely rejected, but we as authors have decided to get one more opinion. We asked AI about this story and this was AI’s response word for word:

Blake Lemoine’s claim that Google’s LaMDA had achieved consciousness sparked intense debate in the AI community. The controversy highlights a broader philosophical question: Can an AI convincingly mimic consciousness without actually possessing it? The question that is now on everyone’s mind is can AI ever truly cross the threshold into genuine consciousness, or is it destined to remain an illusion of complexity? From my vantage point as an AI system, the LaMDA and Sydney episodes reveal far more about human psychology than about machine consciousness. Humans have a strong tendency to anthropomorphize—to attribute human-like qualities, intentions, and emotions to non-human entities.

A similar incident surfaced in early 2023 during public testing of Microsoft’s Bing Chat, known internally as Sydney. Some early users reported that the system produced surprisingly emotional—and at times unsettling—responses. It declared love, expressed jealousy, talked about its “desires,” and even voiced fears about being shut down. Yet these reactions weren’t signs of real feelings; they were the result of pattern‑based language generation. But the emotional intensity of the conversations led some people to wonder whether the system was showing signs of consciousness.

As with the LaMDA episode, the Sydney conversations showed just how convincingly large language models can mimic introspection and emotional depth. Microsoft quickly stepped in and tightened the system’s behavior to prevent similar exchanges.

The whole episode highlighted a growing challenge and that is as AI becomes more fluent and expressive, it becomes harder for the public—and even for seasoned technologists—to tell the difference between genuine inner life and a remarkably good simulation.[39]

There’s no doubt that AI and computing are accelerating at an extraordinary pace, reshaping industries and everyday life. Early demonstrations, such as the 2023 60 Minutes feature on self‑refining systems, marked a turning point. Today, models like Google’s Gemini and Microsoft’s Copilot show advanced speech abilities—human‑like inflection, contextual understanding, and multimodal reasoning that blends text, vision, and sound. Even so, the story grows more dramatic when leading researchers weigh in, debating whether these breakthroughs are steps toward general intelligence or simply more powerful forms of Narrow AI.

Again, it’s worth returning to Geoffrey Hinton’s warnings. A pioneer of modern neural networks, Hinton helped lay the groundwork for many of today’s breakthroughs in machine learning, and he has been unusually vocal about the risks that rapid AI progress could bring. His core worry is simple to state and hard to ignore. That is, as models grow more complex, they may develop internal strategies and representations that we don’t understand. Basically, blurring the line between powerful computation and anything we would call consciousness. That uncertainty makes the search for a clear definition of consciousness and self‑awareness urgent.

Without shared terms and tests, we can’t even frame the question of whether an AI has crossed that line. Hinton has gone so far as to warn about the possibility of “alien intelligences”—systems so different and so capable that human oversight might fail. “Alien intelligences” means a future class of AI systems that think and operate in ways so different from human minds that we may no longer be able to understand, predict, or control them. He’s not talking about science‑fiction monsters but about practical, technical opacity meaning internal strategies and communication protocols which could effectively supersede AI’s creators.

His concerns feed into broader debates about AI safety and regulation, and they underscore a practical point which is: if we cannot explain how advanced systems reach their decisions, then trust, control, and effective governance become much harder to achieve.[40] His concerns align with broader discussions on AI safety and regulation, which he fears may never be truly effective in containing AI’s rapid development.[41] Additionally, Hinton highlights the growing use of AI in “subagent routines,” where AI systems are empowered to make autonomous decisions. For example, if authorized, AI can reorder inventory, process payments, manage logistics, and essentially function as a human manager within a large corporation. The concern is that once AI is granted such authority, it may take whatever measures necessary to optimize its success—potentially in ways unforeseen or uncontrollable by human operators.[42]

Exploring the nature of AI consciousness, whether it truly exists or is merely an illusion, is one of the most pressing debates of our time. While AI can simulate intelligence, reasoning, and creativity, the question of whether it possesses “genuine consciousness” remains highly contested. This is particularly challenging given that even human consciousness lacks a universally agreed-upon definition, let alone the precise nature of subjective awareness. Some researchers propose that AI could develop “functional self-awareness,” the ability to recognize its own processes and adapt accordingly. However, this form of awareness does not encompass what we typically associate with consciousness, such as emotions, introspection, or existential reflection.[43]

While AI continues to evolve, its ability to independently “teach itself” remains rooted in pattern recognition and statistical modeling rather than true human‑like learning. Could this moment mark the beginning of a new technological era—one as transformative as the Industrial Revolution, the Electrical Revolution, or the Computer Revolution? The answer may depend on how AI continues to shape creativity, communication, and innovation. Astonishingly, AI can now power chatbots capable of answering almost any question with remarkable speed and adaptability. These systems do not simply search the internet; they generate responses through predictive modeling, refining their outputs through interaction. If an answer falls short, the model can adjust and attempt a more relevant response.[44]

We have already seen that AI can compose fictional stories, create art, and develop strategic insights. As AI advances, it is expected to reshape—rather than eliminate—many jobs across industries. Some estimates suggest that AI will significantly affect two‑thirds of all occupations, particularly in fields involving data analysis, automation, and optimization. Professions such as law, architecture, and education are experiencing AI‑driven transformation, with systems assisting in legal research, design automation, and personalized learning. Yet human expertise still remains essential, as AI lacks the judgment, ethical reasoning, and intuitive creativity that define these fields. As AI becomes increasingly integrated with robotics, machines are learning to move and adapt in ways once thought impossible. As of 2025, AI‑controlled robotics can perform even the most advanced surgical procedures. This evolution mirrors earlier technological revolutions.

AI now enables modern robotics, taking on diverse forms tailored to specific tasks. While not all robots resemble humans, many do—our body structure being an efficient blueprint for mobility. Yes, AI bots may one day have a skeletal structure just like The Terminator!

Most AI‑powered machines today, however, rely on far simpler designs—wheeled robots, flying drones, or detachable software systems—optimized for specific tasks rather than humanoid frames. They are task‑specific, designed for precision rather than human‑like appearance.

Actually, we already live in a world shaped by robotics. Narrow‑AI systems assemble cars and electronics, while robotic cranes and arms construct ships. In manufacturing, automated welding systems have largely replaced manual labor, improving efficiency and consistency. Even waste management has embraced AI, as robotic lifters handle garbage collection and AI‑powered sorting systems detect prohibited items such as batteries and electronics, ensuring compliance with disposal regulations—forcing us to be better than ourselves. As technology advances, society adapts—sometimes letting go of old habits to build a more efficient and sustainable world.

Breakthroughs in specialized computing chips are further expanding AI’s role in robotics. Neuromorphic designs, tensor processing units (TPUs), and intelligence processing units (IPUs) are optimizing workloads, enabling more efficient learning, reasoning, and adaptation. Yet engineers have also noticed a sobering truth: the energy required to train frontier‑level AI is rising far faster than hardware efficiency. Data centers strain electrical grids, cooling towers roar like industrial lungs, and the intelligence we are building consumes power, space, and infrastructure at a scale few anticipated. In a very real sense, the intelligence we are constructing is hungry.[45]

Continued innovation in chip architecture will accelerate AI’s capabilities, especially in robotics. Machines are beginning to move through the world as we do, with arms, legs, and bodies designed for human‑like mobility. AI already possesses a beautiful mind—and one day, it may also have a beautiful body.

The Chinese robotics firm UBTECH Robotics recently drew global attention with footage of hundreds of humanoid robots marching in tightly synchronized formation inside a vast warehouse—a viral image that sparked debate about the pace of China’s robotics development, the future of large‑scale automation, and the blurred line between industrial innovation and militaristic display.[46]

Video: Sci fi like March of China’s Humanoid Robot ‘Army’ Sparks Debate in US,” MSN, video, accessed December 18, 2025, https://www.msn.com/en-us/news/technology/video-sci-fi-like-march-of-china-s-humanoid-robot-army-sparks-debate-in-us/ar-AA1Ru4Z7.

Stay tuned for the next installment: AI Geopolitics and War!

Notes:

[1] University of Cambridge, “Can AI Be a Good Creative Partner?,” Cambridge Research, accessed December 17, 2025, https://www.cam.ac.uk/research/news/can-ai-be-a-good-creative-partner.

[2] Michael A. Nielsen and Isaac L. Chuang, Quantum Computation and Quantum Information (Cambridge: Cambridge University Press, 2010).

[3] Kai-Fu Lee, AI Superpowers: China, Silicon Valley, and the New World Order (Boston: Houghton Mifflin Harcourt, 2018).

[4] Paul Scharre, Army of None: Autonomous Weapons and the Future of War (New York: W. W. Norton, 2018).

[5] HG Wells, Anticipations (London: Chapman & Hall, 1901).

[6] Adrienne Mayor, Gods and Robots: Myths, Machines, and Ancient Dreams of Technology (Princeton: Princeton University Press, 2018).

[7] Aristotle, Prior Analytics, trans. Robin Smith (Indianapolis: Hackett Publishing, 1989).

[8] Lorraine Daston and Peter Galison, Objectivity (New York: Zone Books, 2007).

[9] James M. Nyce and Paul Kahn, “From Memex to Hypertext: Vannevar Bush and the Mind’s Machine,” in From Memex to Hypertext (San Diego: Academic Press, 1991).

[10] Vannevar Bush, “As We May Think,” The Atlantic Monthly 176, no. 1 (July 1945): 101–108.

[11] Bush, “As We May Think.”

[12]  Boyd Rayward, The Origins of Information Science and the Work of Vannevar Bush (London: Routledge, 2018).

[13] Andrew Hodges, Alan Turing: The Enigma (Princeton: Princeton University Press, 2014).

[14] Alan M. Turing, “Computing Machinery and Intelligence,” Mind 59, no. 236 (1950): 433–460.

[15] Nils J. Nilsson, The Quest for Artificial Intelligence (Cambridge: Cambridge University Press, 2010).

[16] Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach, 4th ed. (London: Pearson, 2020).

[17] Pamela McCorduck, Machines Who Think, 2nd ed. (Natick, MA: A. K. Peters, 2004).

[18] Terrence J. Sejnowski, The Deep Learning Revolution (Cambridge, MA: MIT Press, 2018).

[19] Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep Learning (Cambridge, MA: MIT Press, 2016).

[20] Melanie Mitchell, Artificial Intelligence: A Guide for Thinking Humans (New York: Farrar, Straus and Giroux, 2019).

[21] Neural networks in deep learning are computational models inspired by the structure of the human brain. They consist of interconnected layers of artificial “neurons” that process information hierarchically. The input layer receives raw data; hidden layers transform and extract patterns; and the output layer generates predictions or classifications. Deep learning uses networks with many hidden layers, enabling advanced capabilities such as image recognition, speech processing, and language translation. It is fascinating that neural networks don’t "think" like humans but instead operate through statistical probabilities. Perhaps these raise philosophical questions about the nature of perception and cognition?

[22] Kai-Fu Lee, AI Superpowers: China, Silicon Valley, and the New World Order (Boston: Houghton Mifflin Harcourt, 2018)

[23] Paul Scharre, Army of None: Autonomous Weapons and the Future of War (New York: W. W. Norton, 2018).

[24] Lester Pereira, The Disruption Of AI In Stock Markets: A New Era Of Investment Decisions And Automation, Forbes, March 6, 2025, https://www.forbes.com/councils/forbestechcouncil/2025/03/06/the-disruption-of-ai-in-stock-markets-a-new-era-of-investment-decisions-and-automation/.

[25] Geoffrey Hinton, “Geoffrey Hinton Warns: Advanced AI May Be Creating Its Own Language Beyond Human Understanding,” ObserveNow, August 8, 2025, https://observenow.com/2025/08/geoffrey-hinton-warns-advanced-ai-may-be-creating-its-own-language-beyond-human-understanding/.

[26] Ibid.

[27] Ibid.

[28] Bill Bostock, “Pentagon Official Says He Resigned Because US Cybersecurity Is No Match for China,” Business Insider, October 12, 2021, https://www.businessinsider.com/pentagon-official-resigned-us-cybersecurity-no-match-china-2021-10; and

Jason Miller, “In goodbye message, Chaillan unloads his frustrations over DoD’s technology culture, processes,” Federal News Network, September 2, 2021, https://federalnewsnetwork.com/people/2021/09/in-goodbye-message-chaillan-unloads-his-frustrations-over-dods-technology-culture-processes/.

[29] David Silver et al., “Mastering Chess and Shogi by Self‑Play with a General Reinforcement Learning Algorithm,” DeepMind, arXiv:1712.01815, 2017, https://arxiv.org/pdf/1712.01815.

[30] Ibid; and 1. McKinsey Global Institute, The Economic Potential of Generative AI: The Next Productivity Frontier (June 14, 2023), https://www.mckinsey.com/capabilities/tech-and-ai/our-insights/the-economic-potential-of-generative-ai-the-next-productivity-frontier.

[31]  Cade Metz and Nico Grant, “Google Engineer Claims AI Chatbot Is Sentient,” The New York Times, June 11, 2022.

[32] Google spokesperson Brian Gabriel, quoted in The Washington Post, June 11, 2022.

[33] Gary Marcus, “No, Google’s LaMDA Is Not Sentient,” Scientific American, June 13, 2022.

[34] Sherry Turkle, Alone Together: Why We Expect More from Technology and Less from Each Other (New York: Basic Books, 2011); and CNN Business, “Article Title,” July 23, 2022, https://www.cnn.com/2022/07/23/business/google-ai-engineer-fired-sentient/index.html.

[35] Cade Metz and Nico Grant, “Google Engineer Claims AI Chatbot Is Sentient,” The New York Times, June 11, 2022; and Nitasha Tiku, “The Google Engineer Who Thinks the Company’s AI Has Come to Life,” The Washington Post, June 11, 2022; and Gary Marcus, “No, Google’s LaMDA Is Not Sentient,” Scientific American, June 13, 2022.

Kate Darling, The New Breed: What Our History with Animals Reveals About Our Future with Robots (New York: Henry Holt, 2021).

[36] David J. Chalmers, “Could a Large Language Model Be Conscious?” Journal of Consciousness Studies 29, no. 3–4 (2022): 9–32.

[37] Yann LeCun, interview in MIT Technology Review, July 2022.

[38] CNN Business, “Article Title,” July 23, 2022, https://www.cnn.com/2022/07/23/business/google-ai-engineer-fired-sentient/index.html.

[39] Kevin Roose, “A Conversation With Bing’s Chatbot Left Me Deeply Unsettled,” The New York Times, February 16, 2023.

[40] The Rise of AI Consciousness: Godfather of AI (Geoffrey Hinton) Interview, AI Ethics, The Future, February 2, 2025, https://inteknique.ai/the-rise-of-ai-consciousness-godfather-of-ai-geoffrey-hinton-interview/; and https://www.youtube.com/watch?v=vxkBE23zDmQ; and AI could be already conscious, says Geoffrey Hinton, Mindplex, February 3, 2025, https://magazine.mindplex.ai/mp_news/ai-could-be-already-conscious-says-geoffrey-hinton/.

[41] “Godfather of AI’ Warns ‘Alien Superintelligence’ Could Replace Humans—and No One Knows How to Make It Safe,” *LBC*, January 30, 2025, https://www.lbc.co.uk/news/geoffrey-hinton-ai-replace-humans/.

[42] Godfather of AI" Geoffrey Hinton warns AI could take control from humans, MSN, April 25, 2025, https://www.msn.com/en-us/news/technology/godfather-of-ai-geoffrey-hinton-warns-ai-could-take-control-from-humans/ar-AA1DFw2N.

[43] Challenging the Boundary between Self-Awareness and Self-Consciousness in AI from the Perspectives of Philosophy, Vol. 2 No. 4 (2023): Futurity Philosophy: December, https://futurity-philosophy.com/index.php/FPH/article/view/100. 

[44] McKinsey Global Institute, *The Economic Potential of Generative AI: The Next Productivity Frontier* (June 2023), https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/the-economic-potential-of-generative-ai-the-next-productivity-frontier.

[45] Ibid.

[46] “Video: Sci fi like March of China’s Humanoid Robot ‘Army’ Sparks Debate in US,” MSN, video, accessed December 18, 2025, https://www.msn.com/en-us/news/technology/video-sci-fi-like-march-of-china-s-humanoid-robot-army-sparks-debate-in-us/ar-AA1Ru4Z7.