Consider me ‘concerned about AI’

So, it’s the year 2025, and we’re all falling headlong over ourselves to implement as much AI into our lives as we can.

I’m not too sure that this is a good thing.

Forget the beguiling nature of AI, the fact that it appears to simulate productivity enhancing features. Forget the fact that everyone with an internet enabled microphone seems to hype up the future benefits of AI. Forget the fact that, in many cases, AI produces wildly inaccurate results in a way that you’d find unforgivable were it a human – “just rephrase the prompt”. Forget all that.

Instead, think about whether we want a world in which we relegate ourselves in relation to AI, and the companies behind it / them.

I’m not too sure that I want to live in this world; a world in which decisions are made by machines which don’t have the same incentives as humans do. We’re told that with training, AI can become aligned with the incentives of humanity. Perhaps it can, but is this what we want? Machines deciding / being put in charge of larger and larger parts of what we humans used to do.

High bitrate audio played by a phone is higher fidelity than a vinyl record, but many people prefer the feel of a record. My old shoes feel better than those brand new sneakers, even though the science would say that I’d be better off with something more modern.

There’s certainly something of the ‘rose-tinted spectacles’ and ‘halcyon days’ to my thoughts here, but that’s just where my head is at. I prefer messy over sterile, chaos over brutal efficiency, humanity over machines.

Clearly, this is not always the case, but it’s the new trend for decisions, endeavour, effort, goals to be set by AI which concerns me. Given that we really don’t know what’s going on with the processes AI uses to arrive at answers, I think we ought to let humans make the decisions.