OpenAI

Resources of OpenAI


By AGI, we mean highly autonomous systems that outperform humans at most economically valuable work.


Planning for AGI and beyond, Blog, February 24, 2023


Embedding#

New and Improved Embedding Model, Blog, December 15, 2022


Alignment (Ethic)#

AI Safety Needs Social Scientists, Distill, February 19, 2019

Lessons learned on language model safety and misuse, Blog, March 3, 2022

Our Approach to Alignment Research, Blog, August 24, 2022

Forecasting Potential Misuses, Blog, January 11, 2023

How should AI systems behave, and who should decide?, Blog, February 16, 2023


Models#

RL#

Learning from Human Preferences, Blog, June 13, 2017

Proximal Policy Optimization, Blog, July 20, 2017 - paper

Learning to Summarize with Human Feedback, September 4, 2020 - paper, code

Language#

ChatGPT: Optimizing Language Models for Dialogue, Blog, November 30, 2022

Aligning Language Models to Follow Instructions, Blog, January 27, 2022 - paper

Improving Language Model Behavior by Training on a Curated Dataset, Blog, June 10, 2021

Multi-modal#

CLIP: Connecting Text and Images, Blog, January 5, 2021 - paper, code


Programming#

Triton, GitHub

Resources#

OpenAI Cookbook