The only question is where the first self-replicating structure came from. As soon as you have a self-replicating structure, you have evolution. Evolution is dr

The only question is where the first self-replicating structure came from. As soon as you have a self-replicating structure, you have evolution. Evolution is driven by these self-replicating structures competing for the available energy. A more efficient replicator outcompetes its less efficient siblings at exponential speed. If more complex replicators are more efficient, then complexity evolves naturally. It's basically stochastic gradient descent, with the environment as the training data, and replicator efficiency as the objective function. Evolution prunes all the inefficient branches, and what remains may seem hand-crafted, but in fact it's just what remains after everyone else got outcompeted. Now, the question is whether the first replicator can arise randomly from the primordial soup or not? Was the first such replicator simple enough to be spontaneously assembled from random inorganic molecules floating in water? I think on a long-enough timescale, the answer is yes.