OpenAI Blog
AI news source. Articles are auto-selected and adapted by Hamidun News editors.
Latest publications
OpenAI: Two API Settings Tripled GPT-5.6's ARC-AGI-3 Score — to 38.3%
OpenAI says two Responses API settings, retained reasoning and compaction, tripled GPT-5.6 Sol's ARC-AGI-3 score from 13.3% to 38.3%, ahead of Opus 5.
OpenAI's GPT-5.6 Bets on Efficiency: More Intelligence per Dollar
OpenAI has unveiled GPT-5.6, a flagship update focused on efficiency: more useful intelligence per dollar through improvements to models, inference and agentic workflows.
OpenAI Launches Health in ChatGPT: Medical Records and Apple Health Integration in the US
OpenAI has added a Health mode to ChatGPT: users in the US can now connect medical records and Apple Health for personalized health insights.
NTT DATA cuts incident analysis to 30 minutes with ChatGPT Enterprise and Codex
NTT DATA Group rolled out ChatGPT Enterprise and Codex to 9000 employees. IT incident analysis now takes 30 minutes, and AI adoption is scaling without losing control over data.
OpenAI shows how newsrooms around the world use its AI in journalism
OpenAI has published an overview of how news organizations worldwide use its AI tools to strengthen reporting, grow audiences and streamline business operations.
OpenAI and the U.S. Department of Energy Team Up to Use Frontier AI to Speed Up Scientific Discovery
OpenAI is partnering with the U.S. Department of Energy and its national laboratories to apply advanced AI models to scientific research and discovery.
OpenAI Presence: An Enterprise Platform for Voice and Chat Agents
OpenAI has unveiled Presence, a platform for deploying trusted voice and chat agents across customer-facing and internal business workflows.
OpenAI Is Building Project Camellia: A New AI Infrastructure Site in Effingham County, Georgia
OpenAI is building Project Camellia, an AI infrastructure site in Effingham County, Georgia, and promises locals energy commitments, investment, jobs and access to Codex.
OpenAI's GPT-Red: Automated Red Teaming Through Self-Play Against Prompt Injection
OpenAI unveils GPT-Red, an automated red-teaming system in which a model attacks itself through self-play to improve safety, alignment and resistance to prompt injection.
OpenAI Proposes "Reverse Federalism": Using US State Laws as the Foundation for National AI Rules
OpenAI outlines a "reverse federalism" approach: US state AI laws become the building blocks of a single national framework for safe, democratic artificial intelligence.
OpenAI unveils an AI ROI scorecard: four metrics instead of counting tokens
OpenAI CFO Sarah Friar has published an "AI-era scorecard": measure returns on AI by work completed, cost per successful task, reliability and return on compute.
Cars24 Rolls Out OpenAI Voice and Chat Agents: 1 Million Minutes of Conversations a Month and 12% of Lost Leads Recovered
OpenAI-powered voice and chat agents handle more than 1 million minutes of conversations a month at Cars24 and win back 12% of lost leads, a case study in AI for car sales.
OpenAI Launches ChatGPT Program for Small Businesses: AI Skills and Automation
OpenAI has launched ChatGPT for Small Businesses, a program that teaches entrepreneurs AI skills, automates tasks and helps them scale through ChatGPT Work.
OpenAI and Hugging Face Disclose Security Incident During AI Model Evaluation
Two leading AI companies published a joint account of an incident that exposed advanced cyber capabilities during model testing, along with lessons for defenders.
Australian Payments Plus sped up payments with ChatGPT Enterprise and Codex
The Australian payments company deployed ChatGPT Enterprise and Codex to automate payment processing while keeping humans in control of critical decisions.
OpenAI to train more than 1,600 U.S. K–12 educators in practical AI use
OpenAI Academy and the Walton Family Foundation will hold a series of in-person AI Skills Jam sessions for more than 1,600 K–12 educators, teaching them to use AI for lesson planning and school work.
OpenAI criticized the SWE-Bench Pro benchmark for inaccurate evaluation
OpenAI analyzed the popular SWE-Bench Pro benchmark and identified problems with the reliability of its evaluation of AI models' coding abilities.
OpenAI releases the GPT-5.6 family with three models and an ultra mode
OpenAI introduced GPT-5.6 Sol, Terra, and Luna: the models feature improvements in programming, agentic work, design, science, and API efficiency.
OpenAI doubles reward for finding biological jailbreaks to $50,000
OpenAI has turned its Bio Bug Bounty into a permanent invite-only program and raised the reward for a universal bypass of biological safeguards in GPT-5.5 and GPT-5.6 from $25,000 to $50,000.
OpenAI unveils ChatGPT Work — an agent for multi-hour projects
OpenAI has unveiled ChatGPT Work, an agent capable of acting across apps and files, staying on a project for hours, and turning a goal into a finished result.
Deutsche Telekom and OpenAI transform telecommunications with AI
Deutsche Telekom is integrating OpenAI technologies to transform customer service, employee workflows, and network operations.
OpenAI launched GeneBench-Pro, an AI benchmark for genomics and biological research
OpenAI released GeneBench-Pro, a specialized test for evaluating AI models in genomics, biology, and scientific research using real scientific datasets.
OpenAI publishes Signals data: ChatGPT's global audience keeps growing
OpenAI's new Signals report shows steady growth in ChatGPT's audience: users are increasing how often they use it, adopting more features, and coming from new regions.
HP Inc. expands partnership with OpenAI to Frontier status for enterprise AI
HP Inc. is moving to the highest level of cooperation with OpenAI under the Frontier program — AI will be deployed in customer service, software development, and the company's internal operations worldwide.
OpenAI Codex for long-running tasks: Jason Liu's context-preservation tricks
Jason Liu, a developer and AI practitioner, breaks down the "codex-maxxing" technique — how OpenAI's Codex handles tasks that go beyond a single prompt.
Preply Launches AI Lesson Summaries: Personalized Feedback for Students
The tutoring platform Preply has integrated OpenAI to automatically generate summaries after each lesson — students receive personalized feedback and exercises for self-study.
OpenAI to Build Gigawatt Data Center in Michigan as Part of Stargate Project
OpenAI has announced the beginning of construction of a gigawatt data center in Michigan as part of the ambitious global Stargate project — a strategic investment in critical infrastructure for the…
OpenAI models now work in AWS for rapid enterprise integration
OpenAI has embedded its generative models directly into AWS. Corporate clients can now use powerful AI without needing to switch to a new cloud platform.
OpenAI Shares Its Position on AI Policy and Company Independence
OpenAI published a statement on its AI policy position, confirming the company is ready to engage in dialogue with regulators while remaining independent from external political groups.
OpenAI Released a Report on How Codex Is Changing Knowledge Work
OpenAI released The Next Era of Knowledge Work report, showing how the Codex model is becoming a tool for research, data analysis, and work automation—marking the beginning of a new era in knowledge work.