Ought

Ought is a machine learning research lab automating open-ended reasoning. Ought is building Elicit, an AI-based research assistant for researchers, forecasters, and policy analysts. 

Researchers teach Elicit how to perform a task by providing a few examples. Elicit then uses GPT-3 to scale that task to thousands of publications, questions, and data sources, beyond human capacity. 

Researchers can use Elicit to help define their research question by: 

• Decomposing an ambiguous research question into key factors and more concrete questions
• Searching over proprietary datasets to find related projects or questions from stakeholders

Researchers also use Elicit in their literature review and data gathering processes to: 

• Find the most relevant think tank publications
• Find relevant datasets 
• Find caption-dataset pairs from relevant publications
• Find relevant questions on forecasting platforms 

Ought is building Elicit to ensure that advances in machine learning translate into improvements in reasoning.