Trustworthy Computing

A better method for identifying overconfident large language models

A smarter way for large language models to think about hard problems

Charting the future of AI, from safer answers to faster thinking

Method prevents an AI model from being overconfident about wrong answers

Reasoning and reliability in AI

Automated system teaches users when to collaborate with an AI assistant

Creating space for the evolution of generative and trustworthy AI

The 2022 IBM Research annual letter

The man who challenges AI every day