Why can’t we create machines that grow, like how plants and animals grow?


Why can’t we create machines that grow, like how plants and animals grow?

In: 0

Living things are *complicated*. Making something that intricate, precise, and sensitive out of raw materials is not something we know how to do. The closest we’ve gotten is making computer processors, and they are simple compared to all the stuff going on inside a cell. We just don’t know enough about how life starts, nor do we have the kind of precision manufacturing required to make it ourselves.

As for why we can’t hijack existing living things to make machines that grow, the simple version is that living things tend to want to do their own thing, rather than whatever task you put them to. Getting a plant to grow to a certain rough shape is hard enough, but getting one to create actual functional mechanical components would be downright impossible.

Technically we can, you just have to program the machine to assemble its own parts. But that’s part of the challenge, plants any animals metabolize their own building blocks either from ingesting food or by photosynthesis.

We have to program each individual action into a machine, as well as the decision-making. That is extremely complex to do, because we have to anticipate how each action interacts both with the existing machine and how it will impact future actions.

Simple machines have a very basic process: start at resting state, do action, return to resting state identical to what it started at. Whether it does one or one thousand actions, each action can be carried out the same way.

If the machine is building itself, you have to code the logic to recognize the changes it has made to itself and change its own process to accommodate those changes. If it were stacking pallets, for example, each subsequent pallet must be lifted higher to clear the growing stack of pallets.

Plants and animals can “cheat” by using enzymes and blood/interstitial fluid to carry them around. Instead of a machine doing the work of building itself, each enzyme/protein/cell can act like a tiny machine. So animal and plant growth isn’t a single, ongoing process, but billions of tiny actions going on all the time that result in the overall organism growing.

This is why “nanobots” or “nanites” or “nanomachines” are a field of study: to try to replicate organic growth in an artificial way. We’re still far off from any sort of implementation because it’s hard to code and build effective, reliable machines that are so small. Additionally, we know plants and animals have a built-in failure state (old age, death); our nanomachines might last a lot longer than we expect, and if they’re doing a runaway growth cycle, that may lead to more problems down the road.

We already have nanobots, called Xenobots in this case, that can self replicate. It’s a far cry from an entire machine being capable but it’s a good start.