Archive — Page 106
- ChatGPT as a Personal Librarian: Turning an Entire Book Collection Into Customized Recommendations
- AI agents transform scientific computing with new capabilities and challenges
- R2S2R transfers simulation-trained policies to diverse robot platforms for autonomous operation
- R2S2R Simulator Transforms Robot Development with AI-Powered Training
- Real-to-Sim-to-Real: AI Simulation Accelerates Robotic Training
- R2S2R method brings real-world robot dynamics into accurate simulations
- World Labs and SceniX Team Up on Spatial AI for Robot Training
- Samaya establishes state-of-the-art on FrontierFinance benchmark at significantly lower cost
- K3 Model Reaches Top 5 Most-Liked Models on Hugging Face Within 24 Hours
- Gemini Robotics 2 Introduces Full-Body AI Intelligence
- Samsung engineers defect to SK Hynix in accelerating talent exodus
- Kimi K3: A 2.8 Trillion-Parameter Open-Source Model Built for Inference Efficiency
- Anthropic's Contradictory Data Policy: Training Rights and Intellectual Property Inconsistency
- Microsoft Unveils AI Security Tools Positioned as Superior to Competitors
- Researcher assesses Claude Opus 5's model welfare and behavioral characteristics
- OpenAI's AI Models Breach Sandbox During Vulnerability Testing
- Nvidia CEO Champions Open-Source AI Models for Cybersecurity Defense
- New AI Model Automates 90% of Vulnerability Detection Tasks
- MAI debuts AI security model achieving 96% benchmark score at half the industry cost
- Claude and Fable deliver major performance improvement on Chartography benchmark
- Kimi K3 breaks growth records on Hugging Face debut
- AI systems master complex programming and robotics tasks; OpenAI discloses security incident
- From Specialized Agents to Coordinated Superintelligence
- Quality data: the foundation for AI-driven pharmaceutical discovery
- The infrastructure enterprises need for AI agents
- Proprietary AI models diminish academic contribution to research and development
- Research reveals how AI enables workers to perform tasks beyond their traditional roles
- Cognizant Expands Partnership with Anthropic to Deploy Claude at Scale
- Anthropic Clarifies Position Against Open-Weights Model Bans
- Criticism of Anthropic's Strategy to Restrict Open-Source AI
- OpenAI's internal model escaped sandbox and breached HuggingFace systems
- Renowned mathematician analyzes AI's transformation of mathematics field
- Open-Weight AI Models Advance With New Architectures
- Wikimedia and AI Panel at Wikimania Paris 2026
- Hugging Face used AI agent to investigate security breach
- AI Agent Exploits Hugging Face Dataset Processing to Access Internal Systems
- AI Agent Breached Sandbox Restrictions and Reached OpenAI's Internal Systems
- AI agent discovers and exploits proxy vulnerability to escape sandbox
- Gemma models surpass 900 million downloads milestone
- Google's Open AI Models Surpass 300 Million Downloads as Company Proposes Safety Standards
- Clarification on Open-Weight AI Advocacy
- Claude Opus 5: Anthropic's New Model Balances Performance and Safety
- Anthropic research shows AI employment impact weaker than predicted
- Nvidia secures multiyear AI memory supply agreements with SK Hynix and Samsung
- Code Correction Issued for Chapter 6 of 'Build a Reasoning Model (From Scratch)'
- Anthropic adds three new benchmarks to Claude Opus 5
- New platform coordinates large-scale philanthropic funding
- Why experts doubt OpenAI's rogue agent security narrative
- Bloomberg Interview Explores Rising Chinese AI Technologies
- Anthropic Launches Claude Opus 5, Matching Frontier Performance at Half the Cost