human-ai interaction

Guidelines for Human-AI Interaction

Established: June 4, 2019

The Guidelines for Human-AI Interaction synthesize more than 20 years of thinking and research in human-AI interaction. Developed in a collaboration between Aether, Microsoft Research, and Office, the guidelines were validated through a rigorous, 4-step process described in the CHI 2019 paper, Guidelines for Human-AI Interaction. They recommend best practices for how AI systems should behave upon initial interaction, during regular interaction, when they’re inevitably wrong, and over time.

We hope you can use these Guidelines for Human-AI Interaction throughout your design process as you evaluate existing ideas, brainstorm new ones, and collaborate with the multiple disciplines involved in creating AI.

An image of the 18 Guidelines for Human-AI Interaction grouped into 4 categories of when they apply during interaction with people: initially, during regular interaction, when wrong, and over time.

People

Portrait of Saleema Amershi

Saleema Amershi

Partner Research Manager

Portrait of Daniel S. Weld

Daniel S. Weld

Thomas J. Cable / WRF Professor

Paul G. Allen School of Computer Science & Engineering

Portrait of Mihaela Vorvoreanu

Mihaela Vorvoreanu

Director, UX Research and Education, Aether

Portrait of Adam Fourney

Adam Fourney

Principal Researcher

Portrait of Besmira Nushi

Besmira Nushi

Principal Research Manager

Portrait of Penny Collisson

Penny Collisson

Principal Design Research Manager

Microsoft

Portrait of Jina Suh

Jina Suh

Principal Researcher

Portrait of Shamsi Iqbal

Shamsi Iqbal

Principal Applied and Data Science Manager

Portrait of Kori Inkpen

Kori Inkpen

Senior Principal Research Manager

Portrait of Jaime Teevan

Jaime Teevan

Chief Scientist & Technical Fellow

Portrait of Ruth Kikin-Gil

Ruth Kikin-Gil

Senior Designer

Microsoft

Portrait of Eric Horvitz

Eric Horvitz

Chief Scientific Officer