Artificial Intelligence & Ethics

Global: Personality Steering Influences Cooperation in LLM Agents, Study FindsA recent arXiv preprint investigates how assigning personality traits to large language model (LLM) agents affects their cooperative behavior in controlled…

Global: Study Quantifies Performance Gains from Limited AI-Assisted Cheating in ChessA new study released on January 8, 2026, investigates how occasional use of powerful chess software influences a player’s performance.…

Global: AI Agents Show Conformity Bias in Multi‑Agent Settings, Raising Security ConcernsResearchers who authored a new arXiv preprint (arXiv:2601.05384) report that artificial intelligence agents operating in multi‑agent environments display a…

Global: AI Bill of Materials Platform Aims to Boost Model Transparency and SecurityResearchers announced a new framework that records the provenance of trained artificial intelligence models, offering a verifiable ledger…

Global: Study Examines Memory Poisoning Threats and Defenses for LLM Agents in HealthcareA team of researchers released a preprint in January 2026 that investigates how large language model (LLM) agents…

USA: CAISI Evaluation Finds DeepSeek AI Models Lag Behind U.S. CounterpartsThe Center for AI Standards and Innovation (CAISI) at the National Institute of Standards and Technology (NIST) released an evaluation…

USA: NIST Allocates $20 Million to Launch AI Centers for Manufacturing and Critical Infrastructure The U.S. Commerce Department announced that its National Institute of Standards and Technology (NIST) will allocate…

Global: New Theoretical Bounds Established for Private and Robust Language Model AlignmentResearchers Wenqian Weng, Yi He, and Xingyu Zhou released a study on December 29, 2025 that derives upper bounds…

Global: Researchers Introduce Risk-Aware Stepwise Alignment to Enhance Language Model SafetyIn a recent arXiv preprint, a team of AI researchers proposes a novel method called Risk‑aware Stepwise Alignment (RSA) to…

Global: InSPO: Unlocking Intrinsic Self-Reflection for LLM Preference OptimizationBackgroundA new algorithm called Intrinsic Self‑reflective Preference Optimization (InSPO) has been introduced to improve alignment of large language models (LLMs) by addressing…