Code that comes to life: Google models the emergence of life in the digital environment

Finally, Google has created something that multiplies faster than fake news on the Internet.

Google researchers conducted an experiment simulating the evolution of code fragments over millions of generations, resulting in «self-replicating» programs. This discovery may shed light on the origin of biological life and has sparked lively discussions in the scientific community, as it potentially demonstrates the mechanisms that could have been involved in the emergence of the first living organisms on Earth.

The experiment was based on using tens of thousands of code snippets written in the Brainfuck programming languageBrainfuck is an esoteric programming language invented by Urban Muller for fun. It consists of eight commands, each of which is written with one character.. This language was chosen because of its simplicity — it allows only two basic operations: incrementing or decrementing a value by 1. The researchers randomly mixed and matched these code fragments in a virtual environment, allowing them to execute, interact with each other, mutate, and overwrite themselves and neighboring elements over millions of simulated generations.

Unexpectedly for the scientists, self-replicating programs appeared during the experiment. They quickly reached the population limit set in the experiment. Moreover, new types of replicators emerged, which sometimes outperformed previous versions.

Ben Laurie, one of the authors of the study, emphasized the uniqueness of the experiment: «Everything was chaotic, and then suddenly: boom, they all became the same». He added: «Nothing magical happened. It’s just that physics took a very long time to work, and it led to very complex things».

Professor Susan Stepney called the experiments that led to the emergence of self-replicating programs a «great achievement». In her opinion, this is a significant step towards understanding the possible ways of life’s origin in an environment other than the standard biological «wet equipment».

It is worth noting that Google’s research is not the first of its kind. Back in 1993, at the University of Michigan, Professor Chris Adami led a project that used Avida software to modeling the emergence and evolution of life. Avida created a virtual world where digital organisms competed for resources, mutated, and evolved, mimicking the processes typical of biological systems.

Source: Pcgamer, Newscientist, Nasa