Ai

• See the whole picture and find the look with Circle to Search Feb 25, 2026 New multi-object image search helps you find more items from one picture at the same time in Circle to

• When we prompt a large language model (LLM) to solve a complex polynomial equation, it does not just return an answer but uses its ‘chain of thought’ to work through a solution •

• : Automated Recycling Identification System for E-Waste Classification Using Deep Learning A • : Automated Recycling Identification System for E-Waste Classification Using Deep L

• Closing the Gap Between Text and Speech Understanding in LLMs Closing the Gap Between Text and Speech Understanding in LLMs AuthorsSantiago Cuervoâ , Skyler Seto, Maureen de Sey

• Constructive Circuit Amplification: Improving Math Reasoning in LLMs via Targeted Sub-Network Updates Constructive Circuit Amplification: Improving Math Reasoning in LLMs via Tar

• Our latest threat report examines how malicious actors combine AI models with websites and social platforms-and what it means for detection and defense

• Reusing Pre-Training Data at Test Time is a Compute Multiplier Reusing Pre-Training Data at Test Time is a Compute Multiplier AuthorsAlex Fangâ , Thomas Voice, Ruoming Pang,

• OpenAI appoints Arvind KC as Chief People Officer to help scale the company, strengthen its culture, and lead how work evolves in the age of AI.

• Deploying Open Source Vision Language Models (VLM) on Jetson Vision-Language Models (VLMs) mark a significant leap in AI by blending visual perception with semantic reasoning. •

• AMUSE: Audio-Visual Benchmark and Alignment Framework for Agentic Multi-Speaker Understanding AMUSE: Audio-Visual Benchmark and Alignment Framework for Agentic Multi-Speaker Unde

• Beyond a Single Extractor: Re-thinking HTML-to-Text Extraction for LLM Pretraining Beyond a Single Extractor: Re-thinking HTML-to-Text Extraction for LLM Pretraining AuthorsJeffr

• depyf: Open the Opaque Box of PyTorch Compiler for Machine Learning Researchers depyf: Open the Opaque Box of PyTorch Compiler for Machine Learning Researchers AuthorsKaichao You

• The Potential of CoT for Reasoning: A Closer Look at Trace Dynamics The Potential of CoT for Reasoning: A Closer Look at Trace Dynamics AuthorsGregor Bachmann, Yichen Jiang, Seye

• This story originally appeared in The Algorithm, our weekly newsletter on AI. • To get stories like this in your inbox first,sign up here. • In January, Nvidia’s Jensen Huang, th

• SWE-bench Verified is increasingly contaminated and mismeasures frontier coding progress. • Our analysis shows flawed tests and training leakage. • We recommend SWE-bench Pro. •

• Introducing Frontier Alliances The limiting factor for seeing value from AI in enterprises isn’t model intelligence, it’s how agents are built and run in their organizations. • W

• Reasoning and planning are the bedrock of intelligent AI systems, enabling them to plan, interact, adapt, and ultimately, operate independently. • At Apple, understanding and adv

• Learning to Evict from Key-Value Cache Learning to Evict from Key-Value Cache AuthorsLuca Moschella, Laura Manduchi, Ozan Sener View publication Copy Bibtex The growing size of L

• We share our AI model’s proof attempts for the First Proof math challenge, testing research-grade reasoning on expert-level problems.

• GGML and llama.cpp join HF to ensure the long-term progress of Local AI +404 We are super happy to announce that GGML, creators of Llama.cpp, are joining HF in order to keep futu

• Train AI models with Unsloth and Hugging Face Jobs for FREE This blog post covers how to use Unsloth and Hugging Face Jobs for fast LLM fine-tuning (specifically LiquidAI/LFM2.5-

• AI-enabled deception now permeates our online lives. • There are the high-profile cases you may easily spot, like when White House officials recentlyshareda manipulated image of

• From Scarcity to Scale: How Synthetic Personas Can Bootstrap Japanese AI Development AI could write the next chapter of Japan’s economic story, with forecasts suggesting the tech

• OpenAI commits $7.5M to The Alignment Project to fund independent AI alignment research, strengthening global efforts to address AGI safety and security risks.

• ‘No technology has me dreaming bigger than AI’ Feb 19, 2026 Today CEO Sundar Pichai gave remarks at the opening ceremony of the AI Impact Summit. • Read a transcript of his speec

• OpenAI for India expands AI access across the country-building local infrastructure, powering enterprises, and advancing workforce skills.

• IBM and UC Berkeley Diagnose Why Enterprise Agents Fail Using IT-Bench and MAST Ayhan Sebin Saurabh Jha Rohan Arora Daby Sow Mert Cemri Melissa Pan Ion Stoica ITBench HF Space IT

• Google DeepMind is calling for the moral behavior of large language models-such as what they do when called on to act as companions, therapists, medical advisors, and so on-to be

• A new way to express yourself: Gemini can now create music Feb 18, 2026 The Gemini app now features our most advanced music generation model Lyria 3, empowering anyone to make 30

• The robots who predict the future Three books unpack our infatuation with prediction, and what we lose when we outsource this task to machines. • To be human is, fundamentally, t

• AI Impact Summit 2026: How we’re partnering to make AI work for everyone Feb 18, 2026 Our last decade of breakthroughs in AI are helping us make progress against some of humanity

• OpenAI and Paradigm introduce EVMbench, a benchmark evaluating AI agents’ ability to detect, patch, and exploit high-severity smart contract vulnerabilities.

• One-Shot Any Web App with Gradio’s gr.HTML Gradio 6 quietly shipped a very powerful feature: gr.HTML now supports custom templates, scoped CSS, and JavaScript interactivity. • Wh

• Unifying Ranking and Generation in Query Auto-Completion via Retrieval-Augmented Generation and Multi-Objective Alignment Unifying Ranking and Generation in Query Auto-Completion

• Our 2026 Responsible AI Progress Report We’re releasing our latest Responsible AI Progress Report to share how we’re applying our AI Principles to the development of our products

• Why a 12-year-old forecasting paper has stood the test of time Amazon Scholar Aravind Srinivasan coauthored a 2014 paper about forecasting civil unrest in Latin America, which wo

• Ferret-UI Lite: Lessons from Building Small On-Device GUI Agents Ferret-UI Lite: Lessons from Building Small On-Device GUI Agents AuthorsZhen Yang, Zi-Yi Dou, Di Feng, Forrest Hu

• Models That Prove Their Own Correctness Models That Prove Their Own Correctness AuthorsNoga Amitâ , Shafi Goldwasserâ , Orr Paradiseâ , Guy N. • Rothblum View publication Copy

• Asynchronous Verified Semantic Caching for Tiered LLM Architectures Asynchronous Verified Semantic Caching for Tiered LLM Architectures AuthorsAsmit Kumar Singh, Haozhe Wang, Lax

• A new preprint shows GPT-5.2 proposing a new formula for a gluon amplitude, later formally proved and verified by OpenAI and academic collaborators.

• Introducing Lockdown Mode and Elevated Risk labels in ChatGPT to help organizations defend against prompt injection and AI-driven data exfiltration.

• How OpenAI built a real-time access system combining rate limits, usage tracking, and credits to power continuous access to Sora and Codex.

• GABRIEL is a new open-source toolkit from OpenAI that uses GPT to turn qualitative text and images into quantitative data, helping social scientists analyze research at scale.

• A Small-Scale System for Autoregressive Program Synthesis Enabling Controlled Experimentation A Small-Scale System for Autoregressive Program Synthesis Enabling Controlled Experi

• Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration AuthorsBruno

• Custom Kernels for All from Codex and Claude tl;dr: We built an agent skill that teaches coding agents how to write production CUDA kernels. • Then we pointed Claude and Codex at

• Faster Rates For Federated Variational Inequalities Faster Rates For Federated Variational Inequalities AuthorsGuanghui Wangâ , Satyen Kale View publication Copy Bibtex In this

• Gemini 3 Deep Think: Advancing science, research and engineering Feb 12, 2026 Our most specialized reasoning mode is now updated to solve modern science, research and engineering

• AI is already making online crimes easier. • It could get much worse. • Some cybersecurity researchers say it’s too early to worry about AI-orchestrated cyberattacks. • Others sa

• GPT‑5.3‑Codex‑Spark is OpenAI’s first real‑time coding model, providing instant code suggestions during development. • Delivers 15× faster code generation compared to previous Co

• What’s next for Chinese open-source AI Chinese open models are spreading fast, from Hugging Face to Silicon Valley. • Here’s why that matters. • MIT Technology Review’s What’s Ne

• Mapping the Design Space of User Experience for Computer Use Agents Mapping the Design Space of User Experience for Computer Use Agents AuthorsRuijia Cheng, Jenny T. • Liangâ ,

• OpenEnv in Practice: Evaluating Tool-Using Agents in Real-World Environments AI agents often perform impressively in controlled research settings, yet struggle when deployed in r

• Trace Length is a Simple Uncertainty Signal in Reasoning Models Trace Length is a Simple Uncertainty Signal in Reasoning Models AuthorsSiddhartha Devicâ , Charlotte Pealeâ ¡, Ar

• Is a secure AI assistant possible? • Experts have made progress in LLM security. • But some doubt AI assistants are ready for prime time. • AI agents are a risky business. • Even

• By Ryan Lopopolo, Member of the Technical Staff

• In September, Alfred Stephen, a freelance software developer in Singapore, purchased a ChatGPT Plus subscription, which costs $20 a month and offers more access to advanced model

• 9 fun questions to try asking Google Photos Feb 10, 2026 The Ask button in Google Photos is available on iOS and Android to eligible users in the U.S. • Here are a few ways you c

• Helping kids and teens learn and grow online on Safer Internet Day Feb 10, 2026 On Safer Internet Day, we’re sharing our latest updates designed to support families, kids and tee

• Parallel Track Transformers: Enabling Fast GPU Inference with Reduced Synchronization Parallel Track Transformers: Enabling Fast GPU Inference with Reduced Synchronization Author