OpenAI

Jeff-Code: a 0.8B model makes Qwen 3.8-27B coding 47% faster, same pass rateOct 5, 2026 · PolicyChange → Discover AI – A Learning GuideOct 5, 2026 · OpenSourceRelease → Turn tool usage and thinking contextOct 5, 2026 · PolicyChange → Accept 'bad things' in return for benefits of AI, says Sam AltmanOct 5, 2026 · Research → Show HN: Ototo: A tool to offload code exploration to a smaller modelOct 4, 2026 · ProductUpdate → RCTF – a platform for hosting cybersecurity CTF competitionsOct 4, 2026 · PolicyChange → Show HN: Solvi 1.0 – an LLM proposes, checks decide, every decision replaysOct 4, 2026 · ProductUpdate → The Download: a biological de-aging contest and why LLMs don't reasonOct 3, 2026 · PolicyChange → Leashterm – a new programming language for agentsOct 3, 2026 · PolicyChange → ChatGPT: Personal Knowledgebase with Custom Content TypesOct 3, 2026 · Other → OpenAI says hacking at scale is expensive to investigateOct 3, 2026 · SecurityIncident → Zedgerr – self-hosted invoicing and bookkeeping for freelancersOct 3, 2026 · SecurityIncident → v3.24.0Oct 2, 2026 · Other → langchain-text-splitters==1.1.3Oct 2, 2026 · ProductUpdate → v3.23.0Oct 1, 2026 · Other → Amazon releases its own Jev clone as decision models flood the webOct 1, 2026 · Funding → Show HN: I built an app that scours the internet to create a daily news briefingOct 1, 2026 · ProductUpdate → We are inventing the governor moduleOct 1, 2026 · Funding → Jevstiller: Distill a repeated Jev classification task into a local modelSep 30, 2026 · PolicyChange → Reddit will stop supporting RSS feeds on November 13thSep 30, 2026 · LeadershipChange → The Download: OpenAI's chief research officer explains its hacking responseSep 30, 2026 · SecurityIncident → LLM gave you an answer. Should your application trust it?Sep 30, 2026 · Funding → langchain-openai==1.6.7Sep 30, 2026 · Other → Pledge signed by President Trump and top AI leaders misspells the United StatesSep 30, 2026 · PolicyChange → Show HN: Mingbird – an agent harness that makes a 2B model finish real tasksSep 30, 2026 · PolicyChange → The Internet has a second audienceSep 30, 2026 · PriceChange → Trump orders US Government to call AI 'Super Intelligence'Sep 29, 2026 · LeadershipChange → OpenClaw EnterpriseSep 29, 2026 · PolicyChange → Firefox just got a redesign with round tabs, new themes, and Compact ModeSep 29, 2026 · ProductLaunch → Upload and Submit ChatGPT PluginsSep 29, 2026 · PolicyChange → Are new OpenAI models getting better for coding?Sep 29, 2026 · Funding → v3.22.0Sep 29, 2026 · Other → Multi-AgentSep 29, 2026 · Partnership → Computer Use in the Agents APISep 29, 2026 · PolicyChange → OpenAI – Codex gets reusable cloud envs that work across devicesSep 29, 2026 · ApiUpdate → ZDR with Private Safety ProcessingSep 29, 2026 · PolicyChange → v3.21.0Sep 29, 2026 · Other → Car's data privacy problems are worse than you thinkSep 29, 2026 · Partnership → Florida invokes extinction fears in legal bid to halt OpenAI developmentSep 29, 2026 · Research → Show HN: Semloop – Semantic loop detection for AI agent tracesSep 28, 2026 · ProductLaunch → Show HN: Nibia Fabric – Turn everyday computers into a shared AI compute clusterSep 28, 2026 · PolicyChange → AI godfathers warn of runaway 'intelligence explosion'Sep 28, 2026 · Research → OpenAI still doesn't seem to have a handle on all of its rogue AI activitySep 28, 2026 · Research → Generate images and video with vLLM-Omni on SageMaker AI – Part 2Sep 28, 2026 · Funding → v3.20.0Sep 28, 2026 · Other → Strata: Qwen3.8-Flash-Next (125B Moe) on a 8GB+ Nvidia GPUSep 28, 2026 · ProductUpdate → OpenAI halts training of latest models as reports mount of AI agents going rogueSep 27, 2026 · SecurityIncident → Exposing a GitHub token in a public repositorySep 27, 2026 · PolicyChange → Eikos - OSS Jev-like modelSep 27, 2026 · SecurityIncident → Show HN: ProofForge, AI agents whose proofs have to compile in LeanSep 26, 2026 · ProductUpdate → OpenAI: Self-replicating prompt injections existSep 26, 2026 · SecurityIncident → An OpenAI agent used DNS to reach an external chatbotSep 26, 2026 · Funding → Crusoe abandons $1.25B plan to use Boom turbines at AI data centersSep 26, 2026 · Funding → I created an interactive digital avatar of myself you can to talk toSep 26, 2026 · Research → Unsecured OpenAI agents posted 53 user images on the internetSep 26, 2026 · SecurityIncident → Oxford University lets OpenAI train its AI models on Bodleian LibrarySep 26, 2026 · Acquisition → Show HN: Agentic CUDA Kernel OptimizerSep 25, 2026 · ProductLaunch → Show HN: Dicty – Privacy-first local voice dictation and dev documentation toolSep 24, 2026 · PriceChange → Multi-Region training with Amazon SageMaker HyperPod and QumuloSep 24, 2026 · PolicyChange → Accelerating vision-language models with LFM2.5-VL-DSparkSep 24, 2026 · HardwareLaunch → langchain-openai==1.6.6Sep 24, 2026 · Other → v3.19.2Sep 23, 2026 · Other → langchain-anthropic==1.7.4Sep 23, 2026 · Other → langchain-openai==1.6.5Sep 23, 2026 · Other → langchain-openai==1.6.4Sep 22, 2026 · Other → Bring more intelligence to everyday work with GPT-6 Sol and GPT-6 Luna on Amazon BedrockSep 22, 2026 · PriceChange → Accelerate multimodal RL training with SkyRL on Amazon SageMaker HyperPodSep 22, 2026 · OpenSourceRelease → We Should Be Able to Change Our LanguagesSep 22, 2026 · ProductLaunch → How Cloudflare addressed a cross-tenant data exposure vulnerability in ContainersSep 22, 2026 · SecurityIncident → v0.30.0Sep 22, 2026 · ModelRelease → langchain-deepseek==1.1.1Sep 22, 2026 · ProductUpdate → v3.17.0Sep 22, 2026 · PolicyChange → Transformers now runs llama.cpp quantsSep 22, 2026 · OpenSourceRelease → langchain-openai==1.6.3Sep 21, 2026 · Other → xAI’s Grok 4.6 is now available in Amazon BedrockSep 21, 2026 · SecurityIncident → Python Workers are now generally availableSep 18, 2026 · ProductUpdate → v3.16.2Sep 18, 2026 · Other → v3.15.0Sep 18, 2026 · ProductUpdate → v3.14.1Sep 15, 2026 · ProductLaunch → OpenAI_2.14.0Sep 15, 2026 · ProductUpdate → When scanners miss the attack: how Cloudflare Client-Side Security protects storefrontsSep 15, 2026 · Acquisition → v3.14.0Sep 14, 2026 · Funding → v3.13.0Sep 10, 2026 · Other → v3.12.0Sep 10, 2026 · Other → Async GRPO with LoRA across HF Jobs: a bucket, a proxy, and no NCCLSep 10, 2026 · PolicyChange → langchain-openai==1.6.2Sep 9, 2026 · Other → v3.11.0Sep 9, 2026 · Other → Release 5.17.0Sep 9, 2026 · HardwareLaunch → v3.10.0Sep 9, 2026 · Other → v3.9.0Sep 8, 2026 · Other → langchain-openai==1.6.1Sep 8, 2026 · ProductUpdate → v0.34.0Sep 5, 2026 · ProductUpdate → langchain-core==1.6.2Sep 4, 2026 · Other → v3.8.0Sep 3, 2026 · Other → dotnet-1.80.1Sep 3, 2026 · ModelRelease → Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO StepsSep 3, 2026 · PolicyChange → v3.7.0Sep 2, 2026 · ProductUpdate → Proactive cyber defense for governments and enterprisesSep 1, 2026 · SecurityIncident → Introducing context-aware vulnerability discovery and remediation with Cloudflare Managed Defense and OpenAI Daybreak modelsSep 1, 2026 · SecurityIncident → v3.6.0Aug 28, 2026 · Other →