Evolution of Humanity's Codex: Language, Math, and now AI
Aadharsh Kannan
Pathfinder and incubator driving business value through technology, from conception to launch.
Human beings initially invented language as a tool to communicate complex phenomena—social structure, norms, customs, survival knowledge, and more. However, the early form of language was limited to oral communication. This imposed severe constraints on how knowledge was transmitted and preserved across generations. The only way to pass down generational knowledge was through oral chanting, much like the Hindu Vedas, which were preserved through rigorous oral traditions of recitation and memorization. Each generation faced the same distribution of intelligence. Still, without a written form, it was nearly impossible for high-IQ individuals to consistently pass on more advanced concepts to future generations. This lack of codification also hindered the possibility of specialized work and collective intellectual advancement.
The Advent of Written Language
The invention of the written language changed everything. Information could now be documented and preserved, breaking free from the limitations of oral transmission. This transformation increased the probability that complex ideas could be communicated across generations and geographies, ensuring that similarly skilled and neurodivergent individuals could build upon the work of others even if separated by centuries. A concrete example is the field of mathematics: Isaac Newton famously said, “If I have seen further, it is by standing on the shoulders of giants,” acknowledging his work built upon the contributions of earlier mathematicians such as Euclid and Archimedes. Centuries later, Albert Einstein used non-Euclidean geometry developed by Riemann to formulate the theory of relativity, demonstrating how mathematical thought persisted and advanced across generations.
This evolution was further accelerated by the printing press, which multiplied the scale of information dissemination and allowed for broader codification of complex phenomena such as law, philosophy, and ideology.
The Power of Collective Fiction
Some of the written ideas, though fictional, served an extraordinary purpose. These "collective fictions," as coined by Yuval Noah Harari in Sapiens: A Brief History of Humankind, enabled large-scale human cooperation. One of the most prominent examples of such a fiction is money. Money, unlike barter, has no intrinsic value but serves as a shared construct that billions of people trust and use daily. Without this collective fiction, complex economies and global trade would not exist. Written language provided a new way to solidify these fictions, allowing humanity to create self-sustaining intricate societal systems.
Mathematics: The Universal Language
While humanity was busy codifying more and more complex ideas through language, we also began to formalize another powerful tool: mathematics—the universal language of the universe. This notational system went beyond human-centric concepts and allowed us to characterize and predict the physical world around us. Physics and chemistry were born out of this realization, bringing about the scientific revolution and the establishment of the scientific methodology. Mathematics soon revealed that it could not only describe the universe but could also predict human behavior in aggregate, leading to the birth of economics as a field. As the physicist Eugene Wigner observed, “The unreasonable effectiveness of mathematics in the natural sciences” pointed to a mysterious and powerful alignment between mathematical abstractions and the fabric of reality. This “unreasonable effectiveness” continues to shape fields beyond the physical sciences, demonstrating mathematics' predictive power even in social phenomena.
领英推荐
Language Meets Mathematics: The Age of AI
In a surprising twist, recent advancements have shown that language itself—a uniquely human creation—can be expressed using mathematics. Large Language Models (LLMs), like OpenAI’s GPT, are nothing more than sophisticated mathematical models that predict the next word in a sequence based on a given context. These models function by calculating probabilities and utilizing complex auto-regressive structures, demonstrating that all human knowledge encoded in language can be transformed into a mathematical formula. The implications are profound: the success of AI models (foundational models) suggests that this structure may also extend to vision, videos, and images—essentially anything created by the language-inventing human brain.
Foundational Models: The New Printing Press Moment
For the first time, we are able to turn these mathematical representations into reusable contracts in the form of large foundational models. Traditional computer science sought to create digital representations of reality, but AI models now enable this at an unimaginable scale. This breakthrough is perhaps the printing press moment of the digital age. Just as the printing press revolutionized the dissemination of knowledge, AI now promises to transform how we generate, access, and utilize information.
However, the future remains unpredictable. To illustrate, when the printing press was invented, many believed it would foster rational thinking and displace the myths that organized society. Instead, it gave rise to new forms of myths and misinformation. Today, resonance chambers and the rapid spread of conspiracy theories are a testament to the unpredictable nature of such technological leaps.
The Path Forward: Leveraging AI Responsibly
Love it or hate it, AI is going to fundamentally change humanity. The real question is: How can we wield it to enhance the experience of fellow human beings and ensure it contributes to the advancement of society? Much like religion once served as a collective fiction to unite people towards a shared goal—ensuring survival and prosperity—AI has the potential to either propel humanity forward or become a tool for division and manipulation. The choice is ours to make.
In conclusion, the power of AI is upon us. If not us, then who? If not now, then when?
(We Are Hiring) Director of Engineering (Ex-Salesforce Sr Eng Manager | Ex-LinkedIn Engineering Manager | Angel Investor | Authorized to Work)
5 个月AI is powerful, but it might be dangerous as well, however, we need to use AI to manager AI in a safe way. ??