AI-Written Critiques Help Humans Notice Flaws | allainews.com

June 13, 2022, 6:20 p.m. | Jan Leike

OpenAI Blog openai.com

Showing model-generated critical comments to humans helps them find flaws in summaries.

ai flaws humans research

More from openai.com / OpenAI Blog

We’re bringing the Financial Times’ world-class journalism to ChatGPT 1 week, 2 days ago | openai.com

chatgpt class financial financial times +4

Introducing more enterprise-grade features for API customers 2 weeks, 1 day ago | openai.com

api assistants costs customers +6

OpenAI’s commitment to child safety: adopting safety by design principles 2 weeks, 1 day ago | openai.com

child children commitment companies +7

Introducing OpenAI Japan 3 weeks, 3 days ago | openai.com

asia gpt gpt-4 japan +4

Introducing improvements to the fine-tuning API and expanding our custom models program 1 month ago | openai.com

api build control custom models +6

Start using ChatGPT instantly 1 month, 1 week ago | openai.com

benefits benefits of ai chatgpt experience +2

Navigating the Challenges and Opportunities of Synthetic Voices 1 month, 1 week ago | openai.com

challenges opportunities scale small +4

Sora: First Impressions 1 month, 1 week ago | openai.com

community creative feedback impressions +1

Global news partnerships: Le Monde and Prisa Media 1 month, 3 weeks ago | openai.com

chatgpt french global international +4

Lead Developer (AI)

@ Cere Network | San Francisco, US

View on ai-jobs.net

Research Engineer

@ Allora Labs | Remote

View on ai-jobs.net

Ecosystem Manager

@ Allora Labs | Remote

View on ai-jobs.net

Founding AI Engineer, Agents

@ Occam AI | New York

View on ai-jobs.net

AI Engineer Intern, Agents

@ Occam AI | US

View on ai-jobs.net

AI Research Scientist

@ Vara | Berlin, Germany and Remote

View on ai-jobs.net