AI ethics on Wednesday, 14 January 2026

Abstract The widespread adoption of large language models (LLMs) raises important questions about their safety and alignment 1 . Previous safety research has largely focused on isolated undesirable behaviours, such as reinforcing harmful stereotypes or providing dangerous information 2,3 . Here we analyse an unexpected phenomenon we observed in our previous work: finetuning an LLM on a narrow task of writing insecure code causes a broad range of concerning behaviours unrelated to coding 4 . For

Despite decades of study, large parts of the mammalian metabolome remain unexplored1. Mass spectrometry-based metabolomics routinely detects thousands of small molecule-associated peaks in human tissues and biofluids, but typically only a small fraction of these can be identified, and structure elucidation of novel metabolites remains challenging2–4. Biochemical language models have transformed the interpretation of DNA, RNA and protein sequences, but have not yet had a comparable impact on unde

Introduction: Artificial intelligence (AI) holds profound potential to reshape public health through enhanced disease prediction, diagnosis, and health management. However, this technological advancement is accompanied by significant ethical, social, and governance challenges. This systematic review aims to comprehensively examine the opportunities and challenges of AI in public health, focusing on its applications, associated dilemmas, and governance pathways. Methods: This review was conducted

Purpose: The purpose of this research case study is to examine the Vibhuti Yoga of the Tenth Chapter of the Bhagavad Gita as a philosophical framework for cultivating a sacramental vision that perceives the divine as immanent within the natural and social world. It seeks to interpret how Krishna’s enumeration of divine manifestations provides a non-utilitarian ethical foundation for ecological responsibility, mental well-being, and value-based human conduct. Further, the study aims to contextual

D-amino acid substitution provides an effective strategy for optimizing antimicrobial peptides (AMPs) by enhancing their stability. However, the absence of universal rules renders traditional screening methods time-consuming and labor-intensive, potentially leading to reduced or complete loss of activity. Here, we curated a D-amino acid-substituted AMP dataset from published literature and databases. We then developed ADAPT, an AI-based tool for predicting the functional impact of D-amino acid s