Idea #1

Idea #1 — (The Dangers of) Mimicry

Mimicry has been especially relevant in the AI world lately and it makes sense why. As we create a new form of ever more powerful intelligence we have chosen ourselves and our world as the models. This advancement-by-mimicking works well in the eyes of “progress”, but it raises many issues like, “What will happen when we have fully mimicked ourselves?” and “why not re-tool humans first instead of creating new ones?” and further “is any part of nature private?”. During my time reading Essentialsdigests I’ve come across many great stories on mimicry and I’ll share a bit about one of them here today. It is a thoughtful piece by Wired which has generated further discussion online about the practices we are using to build AI.

Apparently researchers have been using human smell to model and design machine learning algorithms. Turns out human olfaction can help elucidate some of the complications of coding complex systems. The concept of using aspects of nature (in this case human biology) to influence how we design everything from airplanes to machine learning is called biomimicry. This has me thinking about other innate human functions that could be modeled and included. Perhaps sight and touch, or hunger and love could be mapped to better help us understand our world and to help AI better understand it too.

Clearly there is still a long way to go, but I do think that through mimicry we are creating machines that are more like us everyday and we have to talk about it. As AI becomes more distinctly human we must address the question of how it will be treated. Will AI have human rights? Will we limit its access to certain knowledge? As of now we continue our slippery slope downwards towards further uncertainty and at the same time technological advancement. Looking to this new decade it is high time that more thought is put into just what exactly we are trying to accomplish and at what cost. Yes efficiency, profit, and improved quality of life are all worthy goals, but at what point do we assert our dominance over AI? Do we allow AI to at some point become the next evolution of humanity? I’d love to hear your thoughts on the matter. Let’s shape the future together.

Continue reading about Artificial Intelligence
Subscribe to our great newsletter service to stay informed: Register here