That's a good point. I think most people use LLMs by asking questions and receiving answers. But if you reverse the dynamic and have the LLM interview you instead, where you simply respond to its questions, you'll notice something interesting: the LLM as an interviewer is far less "smart" than it is when simply providing answers. I’ve tried it myself, and the interview felt more like interacting with ELIZA [1].
There seemed to be a lack of intent when the LLM was the one asking the questions. This creates a reverse dynamic, where you become the one being "prompted" and this dynamic could be worth studying or adjusting further
>There seemed to be a lack of intent when the LLM was the one asking the questions
There doesn't just seem to be lack of intent, there is no intent, because by the nature of its architecture these systems are just set of weights with a python script attached to them asking you to give you one more token over and over.
There's no needs, drives, motivations, desires or any other part of the cognitive architecture of humans in there that produce genuine intent.
Did you ask the AI to roleplay? All it does is predict what text comes next. Telling it it is role-playing as a perceptive journalist or prominent psychologist should change its predictions.
In some cases using it where you have API access to the system prompt will allow a greater difference in behavior.
That's not an LLM, that's a subscription plan. You can select any OpenAI LLM on ChatGPT Pro.
You can share the chat here, and this will show the LLM you had selected for the conversation. The initial prompt is also pretty important. For claims like current LLMs feel like conversing with Eliza, you are most definitely missing something in how you're going about it.
Advanced voice mode will give you better results for conversations too. It seems to be set up to converse rather than provide answers or perform work for you. No initial prompt, model selection or setup required
There seemed to be a lack of intent when the LLM was the one asking the questions. This creates a reverse dynamic, where you become the one being "prompted" and this dynamic could be worth studying or adjusting further
[1] https://en.wikipedia.org/wiki/ELIZA