Pharmaceutical ads shift AI drug recommendations in controlled study
A controlled study by Mount Sinai researchers found that clearly labeled pharmaceutical ads could shift AI models' treatment preferences, with ...
Tag: AI safety
Study Unpacks How Corporate and Regulatory Language Shapes Perceptions of AI Safety
A comparative cognitive-discursive analysis reveals how tech companies and regulators deploy micro-grammatical choices to construct opposing realities of artificial intelligence ...
NTU Singapore and Google Expand Collaboration on Artificial Intelligence and Scientific Research
Nanyang Technological University and Google have formalized an expanded partnership to advance campus-wide artificial intelligence integration, regional language models, and ...
Position Paper Details Technical and Social Challenges for Responsible Artificial Intelligence
A position paper in AI Perspectives analyzes the technical, legal, and institutional requirements for responsible artificial intelligence, detailing the dual ...
Four As framework proposes new path for governing frontier AI
Anglia Ruskin University researchers propose a Four As framework – Action, Access, Authority and Allocation – for governing frontier AI ...
AI Agents Get Personalities: Scientists Propose Big Five Traits as Design Parameters
Researchers propose treating Big Five personality dimensions as tunable behavioral parameters in multi-agent AI systems, a testable framework rather than ...
Superintelligence Should Be Built Disabled, Position Paper Argues
A position paper argues that beneficial superintelligence should be designed around engineered limitations and an understanding of what AI cannot ...
Low-resource hallucination detection in LLMs on multi-task datasets via iterative pseudo-labeling using confidence thresholding and active learning
Researchers report a semi-supervised framework combining pseudo-labeling, active learning, and ensemble methods that detects fluent hallucinations in large language model ...
Paired response final divergence signatures of jailbreak vulnerability in large language models
A new representational analysis of 15 open-source language models shows that jailbreak vulnerability leaves measurable, family-specific traces in hidden states, ...
Thirteen Rules: Researchers Say the EU and UK Hold the Levers to Rein In AI
A white paper by seventeen researchers argues the EU and UK should use an upcoming summit to adopt thirteen rules ...
-
Mothers who receive childcare support from maternal grandparents show more parental warmth, finds NTU Singapore study
-
University of Seville Breaks 120-Year-Old Mystery, Revises a Key Einstein Concept
-
Bee body mass, pathogens and local climate influence heat tolerance
-
Researchers record first-ever images and data of a shark experiencing a boat strike
-
Groundbreaking Clinical Trial Reveals Lubiprostone Enhances Kidney Function
Embark on a thrilling journey of discovery with Scienmag.com—your ultimate source for cutting-edge breakthroughs. Immerse yourself in a world where curiosity knows no limits and tomorrow’s possibilities become today’s reality!