{"id":15864,"date":"2024-06-17T07:37:03","date_gmt":"2024-06-17T07:37:03","guid":{"rendered":"https:\/\/aurelis.org\/blog\/?p=15864"},"modified":"2024-06-17T07:55:05","modified_gmt":"2024-06-17T07:55:05","slug":"in-the-end-its-the-complexity","status":"publish","type":"post","link":"https:\/\/aurelis.org\/blog\/general-insights\/in-the-end-its-the-complexity","title":{"rendered":"In the End, it\u2019s the Complexity"},"content":{"rendered":"\n

When many elements interact in different ways, they tend to form patterns with some persistence or recurrence. From their interaction, several characteristics sprout, such as life and intelligent behavior.<\/h3>\n\n\n\n

The medium, in its broadest sense, is merely an enabler.<\/p><\/blockquote>\n\n\n\n

Please read first: Complex is not Complicated<\/a>.<\/p><\/blockquote>\n\n\n\n

Across<\/strong> any scale of time or space<\/strong><\/p>\n\n\n\n

For instance, complexity can lead to extraordinary properties not only in physical systems but also in social contexts, reflecting the intricate web of human interactions. Whether large or small, quick or prolonged, centralized or distributed, complexity manifests in various ways. In any of these, there is the potential for \u2018emerging\u2019 intelligence.<\/p>\n\n\n\n

So we reached the e-word.<\/p>\n\n\n\n

Emergence<\/strong><\/p>\n\n\n\n

Complexity fosters emergence. With sufficient drive, emergence becomes inevitable. With enough complexity, emergence leads to intelligence.<\/p>\n\n\n\n

This idea resonates with AURELIS\u2019s focus on the subconceptual level, where complex mental-neuronal patterns<\/a> form the basis of intelligent behavior.<\/p>\n\n\n\n

Moreover, when searching for other intelligences in the universe, we should broadly seek out situations characterized by significant complexity. This search can guide us not only in the cosmos but also in understanding and enhancing artificial intelligence on Earth.<\/p>\n\n\n\n

The amount<\/strong><\/p>\n\n\n\n

A smaller quantity of something contains less complexity than a larger quantity of the same. For instance, this is why the human brain realizes more intelligence than the brains of big apes, as they do relative to those of smaller mammals.<\/p>\n\n\n\n

Elephants and whales also demonstrate intelligence. It\u2019s intriguing to imagine what a Neuralink device linked to advanced effectors might reveal.<\/p>\n\n\n\n

LLMs<\/strong><\/p>\n\n\n\n

Even to the developers\u2019 surprise, large LLMs exhibit intelligent behavior far more effectively than smaller ones.<\/p>\n\n\n\n

It\u2019s not merely the size but the complexity involved. Therefore, achieving the same complexity in a smaller size could yield similar results \u2014 for instance, by eliminating significant redundancy. Then again, what appears redundant may actually contribute to the complexity and resilience of the system.<\/p>\n\n\n\n

Randomness<\/strong><\/p>\n\n\n\n

Intriguingly, there is a sweet spot in this. A certain level of randomness enhances complexity \u2014 or at least its effective behavior. Excessive randomness diminishes it. However, this probably only works indirectly. A certain degree of randomness allows the system to integrate more complexity at another level.<\/p>\n\n\n\n

Once again, it\u2019s the complexity that matters.<\/p>\n\n\n\n

All paths lead to complexity.<\/strong><\/p>\n\n\n\n

We should embrace this scientifically<\/a>.<\/p>\n\n\n\n

\u2015<\/p>\n\n\n\n

Addendum<\/strong><\/p>\n\n\n\n

Me: \u201cLisa, do you think that with increasing complexity, A.I. will increasingly gain intelligence, on and on, endlessly?\u201d<\/p>\n\n\n\n

Lisa: \u201cIncreasing complexity in AI systems does lead to more advanced and sophisticated behaviors, which can be perceived as increasing intelligence. AI systems, like large language models (LLMs), show that with more data and more complex architectures, they can exhibit more intelligent behaviors. However, there are several factors to consider:<\/p>\n\n\n\n