← Back to papers

Paper deep dive

Clarifying AI X-risk

Zac Kenton, Rohin Shah, David Lindner, Vikrant Varma, Victoria Krakovna, Mary Phuong, Ramana Kumar, Elliot Catt

Year: 2022Venue: Alignment ForumArea: Surveys & ReviewsType: PositionEmbeddings: 0

Intelligence

Status: not_run | Model: - | Prompt: - | Confidence: 0%

Entities (0)

No extracted entities yet.

Relation Signals (0)

No relation signals yet.

Cypher Suggestions (0)

No Cypher suggestions yet.

Abstract

DeepMind safety team reviews AI x-risk threat models, proposes a 2D taxonomy (specification gaming vs goal misgeneralization x misaligned power-seeking vs multi-system), and finds more consensus than expected.

Tags

ai-safety (imported, 100%)position (suggested, 88%)surveys-reviews (suggested, 92%)

Links

Full Text

No full-text extraction is stored for this paper yet.