[Project] What if LLM hallucinations were a feature and not a bug?

dreamGPT is the first GPT-based system that uses hallucinations from LLMs for divergent thinking to generate new and novel ideas. Hallucinations are often seen as a negative thing, but what if they could be used for our advantage? We built this autonomous LLM-based agent to try out this hypothesis and the results were quite impressive, The goal of dreamGPT is to explore as many (and diverse) possibilities as possible, as opposed to most other GPT-based platforms which are focused on …

autonomous divergent thinking feature gpt hallucinations ideas llm llm hallucinations llms machinelearning negative novel project thinking