An Inclusive Design Research Centre project

Unsolved Problems in ML Safety

Focus: AI Ethics/Policy

Source: MAIEI

Readability: Expert

Type: Website Article

Open Source: Yes

External URL: https://montrealethics.ai/unsolved-problems-in-ml-safety/

Keywords: N/A

Learn Tags: AI and Machine Learning Ethics Design/Methods Research Centre

Summary: This paper presents four broad areas in ML safety — robustness, monitoring, alignment and systemic safety — and explores each area’s motivations and research directions.