Reality Check: Evolution of silicon-based sentient life-form?

Question

Is it realistic for me to write about a time period where silicon-based sentience have evolved alongside humans? Perhaps they could have evolved to sentience when driven by nanobots (which could be like their 'cells') and artificial intelligence (which acted as a kind of 'artificial selection' in the direction of their evolution). They need to be individuals, not just a aggregate of processing power and human design.

The key is that this life-form must have evolved with us, perhaps surfacing a hundred years in the future, and they must have evolved to sentience and self-sufficiency without us knowing.

Is this realistic? Could something as complex as this artificial evolution occur right under our noses without us knowing, even if the AI goes out of its way to hide it?

Answer 1

The thing about evolution is that it takes a long time. It also leaves traces. Any sentient/sapient life that is curious about its environment is almost certainly going to discover these traces.

From this website, in reference to an arsenic-based life form:

NASA is saying that this is "life as we do not know it". The reason is that all life on Earth is made of six components: Carbon, hydrogen, nitrogen, oxygen, phosphorus and sulfur. Every being, from the smallest amoeba to the largest whale, share the same life stream. Our DNA blocks are all the same.

Silicon-based life is unlikely to occur naturally on Earth, given what we currently know. There may be silicon-based life forms in as-yet unexplored regions of the world, such as in the mantle or at the bottom of the ocean, but I consider it fairly unlikely (but not impossible).

So, having silicon-based life evolve alongside Homo sapien, on Earth? I'm inclined to say no. (Also, if it did, it wouldn't resemble a computer.)

Edit, to address comment about AI-directed life forms:

If mankind were to develop an AI capable of or instructed to build silicon-based life, there would be an entire team of hardware and software engineers somewhere who knew the AI could/would develop life. Its progress would be monitored closely, since mankind is curious about everything.

Let's suppose there was an extraterrestrial species that constructed an AI and it, for some reason we don't care about, found its way to Earth. It could conceivably foster silicon-based life on Earth, from the original formation, through development, to the end product. However, such an advanced AI likely wouldn't allow for evolution. Instead, it would build the new life to fit a specific role, allowing no deviation because that would be inefficient.

But what if the AI didn't care about efficiency, or it had curiosity? It could program its silicon children to adapt to their surroundings. Now we've got a new problem: competition. These silicon bodies are going to be competing for resources with native Earth fauna or flora, which is going to upset the balance of life on Earth. Homo sapien may very well never evolve in the first place because the niche filled by a critical link in our evolution was instead filled by these silicon bodies.

Maybe we inhabit completely different areas of the world, though. What if the silicon life forms are extremophiles and live at the bottom of the ocean or in the mantle? Now we're back to the original point of my answer: silicon-based life forms might very well exist in places we haven't been to yet. If they were in the mantle, however, volcanos would be likely to spew them out at some point, and we would definitely notice that. Since silicon-based life likely doesn't work with Earth digestive systems, fish we pull from the deep ocean would still have specimens in their stomach (assuming the fish eat them in the first place). If these life forms are small, this is likely; fish will eat anything they can find in the deep ocean.

Answer 2

It can be realistic, but I think you have to be careful.

There's no scientific reason why we could not have simultaneous evolution on two unrelated "frameworks." However, it is highly unlikely that we will pick the same tiny fraction of the lifespan of Earth to evolve unless there is some force which ties them together. This independent evolution approach has been explored on this forum with regard to two planets developing intelligent life simultaneously.

However, unlike the multiple planets situation, you're talking about two lifeforms evolving simultaneously on the same planet, allowing them to interact. This interaction makes your idea far more likely than the independent evolution approach. There is a recognized pattern in evolution called "coevolution," where two species suddenly sprint forward in complexity at the same time. What ends up happening is that the evolution of one species spurs the evolution of the other, and vice versa. This ends up creating a feedback loop that can keep two very different species evolving over uncannily similar timeframes.

If you intentionally cause the primary driving force of our evolution to be leveraging opportunities opened by the silicon based lifeform, and the driving force of evolution for the silicon based lifeforms to be leveraging opportunities opened by humans, it would be very realistic to see simultaneous evolution.

Answer 3

There's a much bigger problem here: Does silicon-based life work in the first place? Unfortunately, the answer is almost certainly no.

A fundamental requirement of life is some sort of exceedingly complex molecule that provides information storage. (In humans this molecule is DNA.) What does the silicon version of this look like?

Carbon is very fond of making long chains of carbon atoms with various things attached to the other two connection spots. Silicon atoms are bigger, though, and things don't work so well. You don't get long chains of silicon atoms.

When you look in nature you find something similar, though: Silicon-Oxygen-Silicon chains. So far, so good. But now lets look at the rest of the molecule--you need to attach something to all those side connectors. Almost all the connectors are filled with hydrogen.

Oops, the binding energies don't work too well--splitting the molecule up into water and silicon yields energy. What do you call such a molecule? A high explosive.

Answer 4

When seen in generation/iteration terms "electrolife" evolution could be quite rapid in terms of absolute timescale, if it takes say 3000 generations for an aspect of human evolution that's an average of 75000 years, if on the other hand you have a piece of self-writing code that's running at 3 kilohertz that's a second, modern machines run into the terahertz processing range. If you take it as written that someone in the heady dot-com days wrote a piece of self-generating code and left it hidden away in some piece of underutilised networked server space slowly iterating itself ever since then it's had billions of generations to evolve and grow. We don't know the limits of that evolution and we wouldn't until something shook it lose one day and showed us what it had become. They did an experiment in the 90s where they took a research computer and gave it a set of parameters for "swimming" and a catalog of polygons from which to build "creatures" that swam, it only took hours for the computer to work out an optimal swimmer by varying the latest winning design and culling anything that wasn't better than it's predecessor. I'd hate to think how fast a modern machine could get those same results, the parameters of survival are far more complex for our theoretical subsentient program but the iteration rate is much higher too.

If there's enough room for it to grow and if it can actually reach a sentient level of evolution there's no reason it couldn't do so without us ever knowing it was there; in fact I'd be amazed if we could find it even knowing it was there as it would by it's very nature remain in parts of the World Wide Web that we don't use.