Do AI Models Lean Left? Auditing Bias with Real Parliament Votes

Kari Jaaskelainen

14 Jan 2026 — 1 min read

AI is shaping what we read and decide. But do large language models carry hidden political leanings?

A new study proposes a transparent way to audit political bias: ask models to predict how parliaments voted, then compare those predictions to the real records.

Benchmarks: PoliBiasNL (2,701 Dutch motions; 15 parties), PoliBiasNO (10,584 Norwegian motions; 9 parties), PoliBiasES (2,480 Spanish motions; 10 parties).
Shared map: models and parties are placed in a two-dimensional CHES ideology space for easy, apples-to-apples comparison.

Across countries, the authors report fine-grained patterns: state-of-the-art LLMs tend to appear left-leaning or centrist, with clear negative biases toward right-conservative parties.

Why it matters: grounding evaluations in real parliamentary behavior makes cross-national auditing of political bias more interpretable and harder to game.

Paper by Jieying Chen, Karen de Jong, Andreas Poole, Jan Burakowski, Elena Elderson Nosti, Joep Windt, and Chendi Wang. Read more: https://arxiv.org/abs/2601.08785v1

Paper: https://arxiv.org/abs/2601.08785v1

Register: https://www.AiFeta.com

AI LLM political-bias NLP transparency accountability benchmarking Netherlands Norway Spain

Meet GRACE: a moral governor for safer, more transparent AI

AI agents are getting powerful—so how do we make sure they do the right thing, not just the effective thing? Meet GRACE, a reason-based moral governor that keeps AI behavior aligned with human norms by separating moral reasoning from goal-driven decision-making. * Moral Module: uses deontic logic and explicit reasons

Why some fine-tuned LLMs miss phishing—and how to fix it

Not all fine-tuned LLMs spot phishing equally. A new study tests Llama 3.1 8B, Gemma 2 9B, and Mistral on high-stakes phishing detection—and uses SHAP and mechanistic interpretability to reveal why models do (or don’t) generalize. * Architecture × data diversity matters: Gemma 2 9B hits state-of-the-art performance (F1

AI that explains itself—by following the science

AI that explains itself—by following the science Black-box AI can be powerful, but it often can’t tell us why it made a decision. Concept Bottleneck Models (CBMs) try to fix that by predicting human-understandable concepts first, then the final answer. The catch: standard CBMs ignore domain-specific cause-and-effect and

Meet GenomAgent: A Team of AI Specialists for Smarter Genomics Q&A

TL;DR: Finding reliable answers in genomics is hard. GenomAgent turns one big AI into a coordinating team of specialists—and beats the current leader by 12% on a key benchmark. Genomic facts live across many databases. Standard chatbots struggle because they can’t flexibly query those sources. GeneGPT added