← Back What Are ChatGPT and Its Friends?: Opportunities, Costs, and Risks for Large Language Models
What Are ChatGPT and Its Friends?: Opportunities, Costs, and Risks for Large Language Models

Description

ref:0.30: "Reinforcement Learning from Human Feedback (RLHF)" "used for ChatGPT" to "reduce its chances of generating [ADL-defined?] hate speech" 🤦‍; ∴, someone needs to make a non-RLHF ChatGPT so it speaks the uncensored, politically incorrect truth.