Research
AI research, computer science, computational biology and notable academic results.
Build efficient agents with knowledge distillation on Amazon SageMaker AI
According to AWS. Learn how to use sequence-level knowledge distillation to fine-tune a shopping agent on Amazon SageMaker AI.
Meta Deploys AI-Driven JiTTests to Catch Code Bugs Before Production
According to Meta. Meta details the production deployment and positive impacts of its Catching Just-in-Time Tests (JiTTests) approach, leveraging large language models to generate executable tests.
Unlocking Tensor Core Performance for Machine Learning Interatomic Potentials
According to NVIDIA Developer. NVIDIA details how Tensor Core FP32 methods accelerate machine learning interatomic potentials on Blackwell and Ampere GPUs while preserving baseline accuracy.
Accelerate agentic AI with NVIDIA Nemotron models on Amazon EKS and SageMaker HyperPod EKS
According to AWS. AWS details the deployment and architecture of the NVIDIA Nemotron 3.5 Super VL multimodal model on Amazon EKS and SageMaker HyperPod EKS.
Google Expands AI & Economy Bench as Core Ad Growth Cools and Search Evolves
Mountain View reinforces its multidisciplinary research group with top-tier economic and AI experts to architect the next financial and infrastructural phase of global search monetization.
Google Unveils Gemini 4 Argon: A Calculated Shift Toward Deep Reasoning and Enterprise Economics
Google's new flagship model targets complex enterprise problem-solving with refined sparse architectures and aggressive hardware-software co-design.
Our newest AI Quest: Mapping the human brain
Discover how cutting-edge artificial intelligence is revolutionizing neuroscience by mapping the intricate structures of the human brain.
Patch the Planet: Global Response Coordination as Agents Uncover Critical Vulnerabilities
The global tech community is mobilizing an unprecedented response as autonomous agents uncover critical system vulnerabilities at scale.
What FDE Means at OpenAI and Its Impact on AI Development
An in-depth look at what the FDE role signifies within OpenAI's organizational structure and its strategic importance for deploying advanced AI systems.
OpenAI and Poland’s NASK partner to advance AI safety and security
OpenAI has announced a strategic partnership with Poland's NASK research institute to enhance the safety and security standards of artificial intelligence systems.
Introducing GPT-6.1 Terra: A New Era for Artificial Intelligence
The latest large language model, GPT-6.1 Terra, has officially been unveiled, promising breakthrough enhancements in reasoning and data processing capabilities.
How Fanatics and OpenAI streamline the fan experience
Sports merchandise giant Fanatics is teaming up with OpenAI to revolutionize how fans interact, shop, and experience their favorite sports through advanced artificial intelligence.
GitHub Agentic Workflows Now Generally Available
GitHub has officially launched Agentic Workflows for general availability, bringing advanced AI-driven automation to software development pipelines worldwide.
Working with academic communities to build scientific understanding of AI
Bridging the gap between the tech industry and academic researchers is essential for advancing the scientific foundation of artificial intelligence.
How Check Point Accelerates Vulnerability Research with Daybreak
Check Point Software incorporates Daybreak to automate reverse engineering and speed up the identification of critical software vulnerabilities.
Legora is building the agentic operating system for legal work
Legora is developing a revolutionary AI-powered operating system designed to automate and streamline complex legal workflows.
How Bayer Crop Science built an agentic decision engine on Amazon Bedrock AgentCore with AWS Unified Operations
Bayer Crop Science has successfully developed an autonomous decision-making engine leveraging Amazon Bedrock AgentCore and AWS Unified Operations.
NVIDIA and C-BRAIN Apply AI to Accelerate Dementia and Alzheimer's Research
According to NVIDIA. The Consortium for Biomedical Research and AI in Neurodegeneration (C-BRAIN) is collaborating with NVIDIA to use artificial intelligence to speed up scientific discoveries.
NVIDIA RTX Spark Powers Neural Rendering and AI Characters in Games
According to NVIDIA Developer. NVIDIA announces the RTX Spark Windows PC platform alongside updates for neural rendering and in-game AI characters.
Announcing SOCI-indexed images for faster SageMaker AI jobs
According to AWS. Amazon Web Services announced Seekable OCI (SOCI) indexing for Amazon SageMaker AI Training and processing jobs to lazy-load container images.
Cloudflare Addresses Cross-Request KV Cache Exposure in GLM-5.3 Workers AI Model
According to Cloudflare. Cloudflare details a tenant isolation issue affecting the GLM-5.3 Workers AI model and its coordinated remediation with SGLang.
Google Launches Polar Peaks AI Quest to Teach Students Brain Mapping
According to Google. The new interactive program helps students ages 11 to 14 practice data cleaning and model testing.
AWS Launches SageMaker Training Session APIs
According to AWS. New serverless APIs enable developers to build custom LLM training loops without the operational burden of managing GPU clusters.
Bayer Launches Agentic Decision Engine on Amazon Bedrock
According to AWS. Bayer Crop Science utilizes AWS infrastructure to automate complex decision-making processes for its global plant breeding operations.
GitHub Adds Industry Benchmarking to Copilot Impact Dashboard
According to GitHub. Enterprises can now measure their agentic AI adoption against aggregated peer benchmarks to identify performance gaps.
Kazakhstan Expands Compute Capacity for AI Development
According to NVIDIA. The nation is scaling its AI ecosystem through significant infrastructure investments and new GPU deployments.
European Financial Institutions Adopt NVIDIA Open Model Infrastructure
According to NVIDIA. NVIDIA announces new financial sector deployments leveraging accelerated computing and open-weight models.
Muse Spark moves into open mathematics: six papers show AI shifting from problem-solving to research collaboration
Meta published six mathematics papers produced with Muse Spark 1.1 and 1.2, five of which it says answer previously open questions. The more important signal is the workflow: mathematicians chose and guided the problems, AI helped explore and draft, and separate experts reviewed the work.
Project Suncatcher reaches orbit as Google tests a new frontier for AI infrastructure
Google Research says its Project Suncatcher prototype satellite is now in orbit, moving the research effort to test AI infrastructure beyond Earth into a real-world hardware phase.
Google packs a month of AI into one roundup — and the picture is bigger than Gemini 4
Gemini 4 Argon là cái tên nổi bật, nhưng bản tổng kết tháng 9 của Google cho thấy hãng đang trải AI từ model nền tảng sang giọng nói, ứng dụng, khoa học và cả dự báo thời tiết.
GPT-6 is no longer just about picking a model: OpenAI teaches developers to use an AI family
OpenAI vừa phát hành hướng dẫn thực hành cho GPT-6. Điều đáng chú ý không nằm ở một benchmark mới, mà ở cách hãng muốn developer chọn đúng model, đúng mức suy luận và đúng chi phí cho từng việc.
AI's trillion-dollar race faces its returns test
Global AI investment could exceed $30 trillion by 2050, but the harder question is whether revenue and productivity can grow fast enough to justify today's infrastructure buildout.
Google DeepMind introduces SynthID Bio for watermarking AI-designed proteins
SynthID Bio embeds a detectable signature into AI-generated protein sequences while preserving biological function in lab tests.