Article
Outside of the Bubble, AI Is Black Mirror
Piotr Migdał
- Outside of the tech bubble, AI is an invasive technology with no opt-out options. Here is a brief story of the contrast between my enthusiasm for using AI to visualize Black Mirror season ratings, and the stark reaction of the Internet.
- No hard feelings there - posting on Reddit and being afraid of being banned is like going to the gym and being afraid of sweating.
- I am aware that there are many ways to become “dumb” by using AI, especially if it involves copy-and-pasting without learning. But I like using it to extend my knowledge, especially for things where I don’t know the search terms in advance
- a significant portion of the population does not view AI merely as a tool. If something does not work, AI is called stupid, subjected to a level of scrutiny and nitpicking not applied to humans. If AI does work, it is seen as even worse.
- For any “disruptive tech”, there is friction. And AI is disruptive on many levels: • It is an “intellectual plastic” – so cheap to create that it floods social media with synthetic filler. It is harder to find quality human-made work. • It allows creating fake content with ease. I need to double-check data - superficial polish is a weaker and weaker indicator of quality. • It takes jobs - both ones people love and ones people need. • It challenges some of our core identities as humans - intellectual, creative, and even spiritual.
- We, who train AI models or create AI-assisted tools, need to realize that if we don’t want to face pitchforks, we need to make it easier to use AI for the benefit of humanity, and harder to misuse it. I don’t have a clear answer how.
- For most people, AI feels like a Black Mirror episode they never signed up for. In their story, we are the villains.