February 25, 2024

AI: Harnessing its Power by Understanding its Essence and Human Consciousness

The year 2023 may be remembered as a pivotal moment in the emergence of artificial intelligence (AI). Its potential for both positive advancements and detrimental consequences has become increasingly apparent. For instance, AI models such as ChatGPT, developed by major tech companies like Google, Amazon, and Meta, have already gained an astonishing number of users and are poised to revolutionize various industries, including education and healthcare.

Nonetheless, concerns about AI’s potential for harm were highlighted by a statement signed by prominent researchers, emphasizing the need to prioritize mitigating the risks associated with AI—comparable to threats posed by pandemics and nuclear war. This led 27 nations, including the UK, US, China, and the EU, to vow cooperation at the inaugural AI Safety Summit held at Bletchley Park, England, to ensure safe and beneficial AI development.

One significant challenge in developing AI lies in aligning the AI models with human values, preferences, and goals. However, the complexity of large-scale AI models presents a formidable obstacle, as their inner workings resemble a black box that is incomprehensible to humans. While efforts to enhance transparency and explainability in AI systems are important, they may struggle to keep pace with the rapid innovations in AI technology.

The complexity and uncertainty surrounding AI have given rise to a wide range of beliefs and predictions about its future impact. Some envision a utopian society, while others fear extinction. There are even those who speculate that artificial general intelligence (AGI) will soon achieve sentience. However, the inherent uncertainties in AI exacerbate the challenge of AI alignment. It is crucial not only to ensure that AI models align with human intentions but also that our beliefs about AI accurately reflect its nature and potential.

Beliefs about AI, whether conscious or unconscious, can influence how users interact with AI and how AI adapts to their needs. Psychological studies have demonstrated the phenomenon of self-fulfilling prophecies, where people’s expectations shape outcomes. Manipulating users’ beliefs about AI can increase the likelihood of various outcomes, including catastrophic ones. Therefore, gaining a deeper understanding of AI’s essence and its implications is essential.

To comprehend AI’s essence, one must understand its origins and its relationship with human cognition. AI, as automated computation run on computers, has roots in the cognitive sciences, neuroscience, and computer science. The convergence of these disciplines has led to the belief that AI, being computational in nature, mirrors the computation of natural intelligence—the human mind.

Computation itself traces back to the foundational efforts to define arithmetic in logic, dating back to mathematicians Richard Dedekind and Giuseppe Peano in the late 19th century. While mathematician Kurt Gödel showed that it was impossible to achieve the goal of grounding all mathematics on a secure formal basis, his work inspired the development of the Turing machine—the underpinning of computer science. Computation, therefore, emerges from the mathematical concepts that were originally designed to define arithmetic.

Recent research by a group of scholars has revealed that arithmetic is deeply intertwined with perception. It is shaped by a fundamental structure that acts as colored glasses, organizing our experience of the world. Arithmetic, consisting of numbers and operations like addition and multiplication, arises from the constraints imposed by this structure. In other words, our perception, influenced by the same underlying structure, gives rise to the concept of numbers and arithmetic.

This understanding showcases the intimate relationship between mathematics and the human mind—a reflection of its nature and creativity. While the deep structure of perception is shared among animals, only humans have developed mathematics, making it a uniquely human creation. Enabling the development of AI, mathematics holds significant consequences for our technological advancements.

This perspective on the origin of arithmetic aligns with the insights of philosopher Immanuel Kant from the 18th century. Kant argued that our knowledge of the world is structured by pure intuitions of space and time, existing prior to sensory experiences. These intuitions can be likened to the colored glasses that shape our perception. Kant’s philosophy, regarded as a Copernican revolution of the mind, challenges the perception that knowledge is solely derived from sense experience.

Recognizing the biological foundation of arithmetic resulting from human perception suggests a paradigm shift in our understanding of computation. Computation is not separate from us in an abstract realm of mathematical truth; it is inherent in the nature of our minds. The brain is not merely a computer; computation—the basis of AI—is an expression, much like mathematics, of the mind’s nature and creative capacity.

By unlocking a deeper understanding of the essence of AI and acknowledging its connection to human consciousness, we can navigate its development more responsibly. Harnessing the power of AI wisely requires aligning AI models with human values and goals while remaining conscious of our own beliefs and their potential impact on AI outcomes. Only by fully grasping the nature of AI and our minds can we ensure a future where AI serves humanity’s best interests.

1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it