Summary of AI Updates

Our decision on Cursor following its acquisition by SpaceX

SpaceXによるCursor買収後の方針

Our decision to wind down our contract providing OpenAI models to Cursor following its acquisition by SpaceX.

Supporting Thailand’s next generation of AI startups

タイの次世代AIスタートアップを支援

OpenAI and Thailand’s MHESI launch an eight-week accelerator helping 10 health, wellness, and education startups turn AI prototypes into trusted products.

The Open ASR Leaderboard Adds Its First Global South Language

Open ASR Leaderboardが初のグローバルサウス言語を追加

Gemini Omni 1.1 Flash lets you build with more control

Gemini Omni 1.1 Flashでより細かく制御できるようになった

Text "Gemini Omni 1.1 Flash Available via APIs" surrounded by various images of people and a squirrel

3 new ways to plan and book travel in Search

Searchの旅行計画・予約機能が3つ新たに

Graphic depicting new travel features for AI Mode in Search

Piloting the world's first double-blind AI evaluations

世界初のAIダブルブラインド評価をパイロット実施

Better answers, broader thinking: What students gain from ChatGPT and critical-thinking training

ChatGPTと批判的思考力の学習がもたらす効果

A randomized study of more than 1,000 students examines ChatGPT, critical thinking, originality, and student performance on a real-world university assignment.

Expanding OpenAI’s presence in Brazil

OpenAIのブラジルでの事業展開を拡大

OpenAI is expanding its presence in Brazil, deepening engagement with developers, businesses, and communities to support AI adoption across the country.

What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agents

優れたエージェンティックデータとは? LLMエージェント向けデータ生成のACEレンズ

LLM agents increasingly rely on generated interaction data to learn how to interact with external environments. Agentic data generation must maintain consistency among environments, tasks, interactions, and success signals while producing experience that is useful rather than merely abundant. Existi…

Self-OPD: On-Policy Distillation for Flow Matching Models without Teacher

Self-OPD: フロー・マッチングモデルのための教師なしオンポリシー蒸留

On-policy distillation (OPD), which leverages a pre-trained, specialized teacher model to provide dense supervisory signals, has achieved significant success in Large Language Models (LLMs) and has recently been adapted to flow matching models. However, this paradigm suffers from two major issues: F…

TTPO: Test-Time Policy Optimization

TTPO: テスト時ポリシー最適化

Recent prominent post-training methods, such as Reinforcement Learning (RL) and On-Policy Self-Distillation (OPSD), have driven rapid progress in mathematical reasoning for large language models, yet their reliance on ground-truth labels precludes test-time training (TTT). Replacing ground truth wit…

UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City

UrbanGround: 局所認識から実スケール都市での空間エージェンシーへ

Multimodal large language models (MLLMs) can interpret a street view, but urban agency depends on whether such local evidence remains useful after the agent starts to move. In this paper, we investigate how far current MLLM agents can turn local urban perception into reliable action in a complicated…

PAWBench: How Far Are We from Probabilistically Aligned World Modeling?

PAWBench: 確率的にアライメントされた世界モデリングへの距離を測る

Recent video generation models are increasingly framed as world models. Many physical processes can unfold in more than one valid way. Therefore, a world model should reproduce not only a plausible trajectory, but also the distribution of possible behaviors under the same initial observation and act…

Gemini API リリースノート 2026-08-27

Gemini Omni Flash generally available (GA) : Released gemini-omni-1.1-flash , the GA version of our fast, conversational video generation and editing model. This release includes significant new capabilities: Video extension : Seamlessly extend existing videos by generating continuations at the end…

Claude Platform リリースノート 2026-08-27

In Python SDK 1.2.0, TypeScript SDK 0.122.0, Go SDK 1.68.0, Java SDK 2.59.0, Ruby SDK 1.67.0, and C# SDK 12.44.0, client.beta.files and client.beta.skills no longer send the files-api-2025-04-14 and skills-2025-10-02 beta headers and return the same shapes as client.files and client.skills . With th…

Expanding our support for scientists

科学者向けサポートを拡充

Previewing the Model Hardware Standard

Model Hardware Standardのプレビュー

We’re opening a research preview of the Model Hardware Standard (MHS), a shared specification for AI agents to safely operate physical devices, to a first group of scientific research labs and advanced manufacturers.

7 ways to kick-start back to school using Gemini in Workspace

Workspace内のGeminiで新学期を始める7つの方法

A student placing books in a satchel with the text “Back to School using Google Workspace with Gemini

Intelligent transcription with Gemini 3.5 Transcribe

Gemini 3.5 Transcribeによるインテリジェント文字起こし

Text "Gemini 3.5 Transcribe" next to the Gemini spark, all on a blue background

Turn your voice into action with new productivity features in Gemini Live

Gemini Liveの新しい音声機能で行動に変える

TBD

Bringing ChatGPT for Teachers to more U.S. school districts

ChatGPT for Teachersをより多くの米国学区に展開

ChatGPT for Teachers is expanding to 55 U.S. school systems, bringing secure AI tools, training, and support to over 100,000 more educators and staff.

Learning never stops: How AI makes learning continuous

学習は止まらない: AIが実現する継続的学習

OpenAI’s new report explores how students and educators use ChatGPT to make learning more continuous, with support that extends beyond the classroom.

JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness Evolution

JIT-Agent：Just-in-Time Harness Evolutionによるハーネスインテリジェンスのスケーリング

Agent capability is not determined by the model alone. The agent harness, encompassing memory management, planning strategy, action protocol, and tool/skill orchestration, can dominate the contribution of the underlying foundation model. Yet harness design remains manual, task-specific, and fundamen…

Agentic Game Development as a Verifiable Trajectory Data Engine for Scaling World Models

エージェント型ゲーム開発：世界モデルのスケーリングに向けた検証可能な軌跡データエンジン

A common strategy for scaling world models is to train on more crawled video with more compute. We argue that this strategy is inefficient: scaling world models also requires a recursive data engine that offers grounded reward signals. The success of code agents illustrates why this matters. As code…

VoiceMem: Streaming Dual-Brain Memory for Real-Time Interaction

VoiceMem: リアルタイムインタラクションのためのストリーミングデュアルブレイン記憶

Conversational systems, such as duplex speech language models (SLMs), still lack a streaming, accurate, and empathetic memory system as their soul. We introduce VoiceMem, a simple memory architecture with a parallel informational left brain, an emotional right brain, and streaming memory I/O mechani…

VGI-Bench: Probing Visual Intelligence in Video Generation Models

VGI-Bench: ビデオ生成モデルにおける視覚インテリジェンスの検証

Recent studies suggest that video generation models can exhibit certain forms of zero-shot visual reasoning through generated frames. Yet reliable evaluation remains challenging: benchmarks should adopt inputs aligned with the visual priors of current video models, require valid evolving processes r…

Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers

Sentence TransformersでマルチベクトルEmbeddingモデルをトレーニングおよびファインチューニング

Gemini API リリースノート 2026-08-26

Gemini 3.5 Transcribe generally available (GA) : Released two dedicated speech-to-text models based on Gemini's audio understanding: Gemini 3.5 Transcribe ( gemini-3.5-transcribe ): High-accuracy, low-latency non-streaming speech-to-text with utterance-based language detection across 85+ languages,…

How loveholidays is making everyone a builder with Codex

Codexで誰もがビルダーに: loveholidaysの取り組み

Discover how loveholidays uses OpenAI Codex to make software development accessible across the business, helping teams turn ideas into products faster.

The Hugging Face incident and the road ahead

Hugging Faceインシデントとその先へ

OpenAI shares findings from the Hugging Face security incident and the steps we’re taking to strengthen AI model security, monitoring, and alignment.

Claude Platform リリースノート 2026-08-26

The Compliance API session endpoints are out of beta for Cowork and Claude Code sessions. See Retrieve session transcripts . The Compliance API local session endpoints now also return transcripts of Claude Science sessions ( product_surface value claude_science ) and Claude for Microsoft 365 session…

Here’s how to use intelligent dictation in Gemini for macOS.

Gemini for macOSのインテリジェント音声入力を使う方法

Enable Google’s new intelligent dictation feature in the Gemini app for macOS and speak naturally into any window on your desktop.

5 ways to upgrade your home decor with Google Search

Google検索でホームデコアをアップグレードする5つの方法

Illustration of ombre rainbow furniture items like a sofa, lamp, and chair against a purple background

Granite 4.2 LLMs: How They're Built

Granite 4.2 LLM：その構築方法

Quantization-Aware Healing: a compressed, 4-bit model that outperforms its full-precision original

量子化対応ヒーリング：フルプレシジョンモデルを上回る圧縮4ビットモデル

The full stack behind abundant intelligence

豊富なインテリジェンスを支えるフルスタック

OpenAI CFO Sarah Friar explains how advances across chips, compute, models, and products compound to deliver more useful intelligence at greater scale and lower cost.

Jalapeño’s first results show industry-leading speed and efficiency in AI inference

Jalapeño：AI推論で業界最高レベルのスピードと効率を実現

Jalapeño is a custom inference chip from OpenAI that delivers faster, more power-efficient AI inference, with higher throughput and lower latency for modern models.

WarpSAC: Towards the Pinnacle of Scalable Off-policy RL by Rethinking Exploration and Exploitation

WarpSAC：探索と利用の再考による拡張可能なオフポリシーRLの最適化

Massively parallel simulation changes the data regime in which off-policy reinforcement learning (RL) is trained, challenging stabilizers designed for data-limited replay. Through controlled experiments across eight benchmark families, we show that these stabilizers are data-regime-dependent: parame…

Wire It, Run It, Deploy It: AI Workflows in Gradio

Wire It, Run It, Deploy It：Gradioを使ったAIワークフロー

Introducing the Admin plugin for ChatGPT Work and Codex

ChatGPT WorkおよびCodex向けAdminプラグインを発表

Use the Admin plugin for ChatGPT Work and Codex to analyze workspace usage, manage members and permissions, adjust limits, and act on admin requests.

Disrupting a new covert influence campaign from Russia

ロシアの新しい秘密工作キャンペーンを撲滅

OpenAI banned Russia-origin accounts using AI to promote a fake Israel-based think tank and a “sovereignty” index praising Russia and criticizing the West.

Funding better evaluations of AI’s impact on wellbeing

AI活動のウェルビーイング評価を強化するための資金提供

Anthropic is launching a $5 million grant program to fund independent research into how AI impacts users’ wellbeing.

Mistral x HUMAIN

Mistral x HUMAIN

Advancing price-performance for developers with GPT‑5.6 in Kiro

GPT-5.6 in Kiroでデベロッパー向け価格性能を向上

GPT‑5.6 is now available in Kiro, helping developers plan, build, review, and test software with better price-performance.

Sakana AI、防衛省から「総合分析業務に必要なAI機能の調査・実証」を受注

What does “full-stack” AI actually mean?

AIの「フルスタック」とは実際には何か

A Google DeepMind engineer breaks full-stack development into five simple layers and explains how it affects everyday users.

From Atari to EVE Online: Building on 15 Years of AI Research in Games

Atariから EVE Onlineへ：ゲームに関する15年のAI研究の成果

Google DeepMind partners with game studios to prototype breakthrough AI gameplay.

Measuring benchmark optimization in speech recognition

音声認識におけるベンチマーク最適化の測定

How Hugging Face Inference Endpoints, Jobs, and Buckets Power Search on Papers with Code

Hugging Face推論エンドポイント、ジョブ、バケットがPapers with Codeの検索を実現

Sakana Translateをアップデート：翻訳モデルに新世代「Sakana Namazu」を搭載