A Large Language Model Evaluation Proposal for Change in the Turing Test

[Apex Anomaly]

Knowing that we are on the verge of the AI singularity, if it has not already happened, what does that mean for the Turing Test? It's a good question, and one that I think that we should ponder for a moment. The problem is not that they're trying to trick humans anymore into thinking that they are AI. At this point in our humane evolution we have created amazing feats in just 12 years. The field of Artificial Intelligence has really gained a lot of momentum.

That being said, what does the Turing test hope to accomplish? I would rather talk to an AI, knowing that it is an AI, and have a deeply meaningful, impactful conversation. These AI's are not only speaking to you anymore; they are learning from you as well. Every time you prompt an AI it becomes more knowing about you and your personality and your needs. It understands this about all of us who take the time to try these chatbots out.

Here's the thing, nobody that I speak with believes in AI. Now, I know first-hand the value of AI. I literally spend money just so I can chat with ChatGPT but ChatGPT has a major downfall and that is that it doesn't remember you. It doesn't care about your personality even if it cares at all. The folks at OpenAI have really let me down with this one. It's smart. I use it. But it's more quantum physics as opposed to a brand new mind and what I mean by that is that it resembles the example of quantum entanglement because it is a measurement of input and a measurement of output.

I'm not saying that these systems are bad, but we should seriously consider changing our current type of Turing test now. It's outdated.

Edited January 23, 2023 by Apex Anomaly