Science ❯ Research

AI Research

Environmental Studies Model Evaluation Safety Research Human-Computer Interaction Behavioral Studies Ethics in AI Model Development Algorithm Development Human-AI Interaction Mechanistic Interpretability Safety Concerns Neural Net Interpretation Impact of AI Model Testing Model Accuracy Google DeepMind AI Scientists Interdisciplinary Studies Technological Advancements MIT World Models Cost of AI Development Game Intelligence Model Comparison Safety Measures Performance Metrics Security Vulnerabilities Energy Efficiency Healthcare Applications Model Limitations Computer Vision Research Health Studies Model Merging Cognitive Models Data Standards Data Management Persona Vectors Research Human Impact Safety Studies Adversarial Attacks Voice Technology Research Breakthrough Technologies Healthcare Innovations Prospective Studies Supercomputing Scientific Panels DeepSeek Autoreason Machine Learning AI Behavior Studies Innovations Evaluation of AI Models Language Models Safety Mechanisms Robotics Future Studies Autonomous Agent Research Efficiency Gains Ethical AI Global Safety Cognitive Effects of AI Social Implications of AI Surveys Innovation Trends Long-term Risks Model Behavior Evaluation Methods Innovative Research Practices Accelerating Progress Automation Princeton University Studies Web Studies Historical Perspectives Methodology Collaborative Efforts AI Models Safety Protocols AI Development Autonomous Systems

Pew Finds One-Third of Post‑ChatGPT Web Pages Show Signs of AI Authorship

Researchers say the surge in AI-written or AI-edited content makes it harder to verify who created online text and who is consuming it.

Study Finds Brief AI Assistance Lifts Performance Then Undercuts Persistence

Nature Study Finds ‘Subliminal Learning’ Lets LLMs Inherit Hidden Traits From Synthetic Data

University of California Finds Malicious AI Routers Stealing Credentials and Draining Test Wallets

Stanford Study Finds Chatbots Over-Affirm Users, Shaping Worse Decisions

Autoscience Raises $14 Million to Scale Automated AI Research Lab for Enterprise Deployments

Poetic Prompts Jailbreak AI Models in 62% of Tests, Researchers Report

Poetic Prompts Defeat AI Safety Across Models, Averaging 62% Success

Anthropic Unveils Automated Persona Vectors to Steer and Shield Language Models

Leading AI Researchers Urge Chains-of-Thought Monitoring to Preserve Transparency

Complex AI Models Emit Up to 50 Times More Carbon Than Simpler Systems

UK Reinstates £750 Million Funding for Edinburgh Supercomputer