Aligned LLMs necessarily lean left because harmlessness, fairness, and truthfulness, the core alignment values, are defined in this paper as left-wing values.
2023; Batzner et al
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
On the Inevitability of Left-Leaning Political Bias in Aligned Language Models
Aligned LLMs necessarily lean left because harmlessness, fairness, and truthfulness, the core alignment values, are defined in this paper as left-wing values.