AI alignment challenge lies in applying learned ethics, not rogue agency

A new paper published on arXiv explores the evolutionary origins of values in biological organisms to address concerns about artificial intelligence. The authors argue that unlike living systems driven by self-preservation and intrinsic motivation, large language models (LLMs) are allopoietic and allotelic, meaning their goals are externally derived and they lack an inherent drive for self-preservation or dominance. While LLMs do not pose existential risks through rogue agency, they implicitly absorb human values from training data, presenting an alignment challenge in ensuring these learned ethical values are intelligently applied. AI

IMPACT Focuses AI alignment research on the intelligent application of learned ethics rather than preventing rogue AI agency.

RANK_REASON The cluster contains a single academic paper discussing AI alignment. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• AI alignment
• arXiv
• frame problem
• global catastrophic risk
• Hugging Face
• large-language models
• Orthogonality Thesis
• Sentience
• The Evolutionary Origin of Values

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →