Skip to content
English

Switch theme

Discover▾

Not sure where to start? This week’s hottest software, models and stacks in three minutes.

AI Apps▾

AI software people use: from assistants like ChatGPT to tools for coding, video and workflows, picked by purpose.

For Agents▾

Software for AI: tools that let your agent (Claude Code, Codex, Cursor…) browse, read docs, run code and remember.

Models▾

Choose models, estimate cost, self-host: prices, rankings, benchmarks, free tiers and compute.

Solutions▾

A goal or an error message? Ready-made stacks, steps and fixes by scenario.

Learn▾

A systematic course from zero to the frontier, with an algorithm library, must-read papers and tutorials.

Frontier▾

Daily AI news, weekly highlights, the latest papers and how AI is changing every discipline.

// Frontier

Products & applications

AI news in products & applications, newest first. Headlines link to the original.

Today only This week’s hotspots →

last updated 2026-10-22 19:00 (Beijing time)575 stories · 185 sourcesoriginal links only

01Timeline

Grouped by day (Beijing time). Headlines stay in their original language.

All Research & conferencesChips & computePolicy & regulationAI safetyProducts & applicationsFunding & companiesAI × humanities & society
More ▾
AI × maths 15AI × biology & medicine 109AI for Science 133AI × finance 136Other 129

Thu, Oct 22 · today61

Eiliko maker says it’s hiring an ‘’independent security firm’ after Verge report on its insecure robots

The Verge AI

ChatGPTやGeminiの性格診断結果をArtificial Analysisが公開、Gemini 4 Argonは媚びやすくClaude Fable 5.1は説得力がある

GIGAZINE

AI時代の言語脆弱性ラッシュにインフラエンジニアはどう立ち向かうべきか？

Qiita AI

Hugging Face’s New AI Biology Model Lets You Chat With DNA

WIRED

+1 more outletsBusiness Insider

TF RAID Avengers: AI swarming competition

GOV.UK

複数のAI画像モデルを使っているとき、どうやって製品画像の一貫性を保っていますか？

Qiita AI

同じGGUFなのにLM Studio 6.2 tok/s → Ollama 19.1 tok/s。VRAM 8GBでQwen3.6-35B-A3Bを実測した

Qiita LLM

MCP サーバーのツール一覧はいくらかかるのか — 34,374 件のトークン数を数えた

Qiita LLM

엔비디아, 로봇 데이터 수집 비용 125배 낮추는 모델 '플로라' 공개

AI Times Korea

미스트랄, 실시간 음성 에이전트 지원 '복스트랄 3' 제품군 공개

AI Times Korea

情報漏洩の次に何が起きるのか？――AI時代のサイバー犯罪を予測し、防御策を考える

Qiita AI

【見逃し厳禁！】AI個人開発で賞金を狙え！総額20,000ドル＆25ドル分クレジットのBeamコンテスト！

Qiita LLM

HarmBench・StrongREJECT で何が測れて何が測れないか：データの中身を数え、自分のアプリとの差を洗い出す

Qiita LLM

Ensemble forecasting with IBM Granite time-series models

Hugging Face Blog

既能“做选择题”，也能“做填空题”——星辰意图理解模型Xing4.0-29B-Intent发布

量子位

谁在给OpenAI供应具身数据？

量子位

AI Trade Jitters Return, Oil Gains After Deadly Saudi Attacks

Bloomberg Politics

豪州でAI「Claude」に保育施設銃撃を予告したら逮捕 弁護側「耳感染症でストレス」

ITmedia NEWS

NVIDIA Cosmos Opens New Tools to Train, Evaluate and Deploy World Models

NVIDIA Blog

NVIDIA Transforms European Retail, From Digital Discovery to Physical Operations

NVIDIA Blog

Pioneers Across Europe Deploy NVIDIA Earth-2 Open Models to Power Real-World Weather and Climate Prediction

NVIDIA Blog

What is PHP and why did it spark a global cyber-security scramble?

BBC

中国拟出台新规，加强对先进人工智能模型出口管控

纽约时报中文网

Canadian publisher Boréal ends relationship with novelist Thélyson Orélien over alleged use of AI

Guardian Tech

A hybrid artificial intelligence framework for three-dimensional electromagnetic field mapping using airborne measurements in general aviation

Nature

AI-quantum hybrid channel modeling and noise mitigation for 5G/6G wireless networks: the AQCME framework

Nature

走向超级智能体，AI还需要主动性：UIUC、谷歌等联合综述

36氪

+1 more outlets新智元

宇树科技王兴兴：具身智能迈过GPT时刻，最大挑战是模型与真实机器人的对齐

36氪

+1 more outlets界面新闻

黄仁勋14天交付算力集群，配合DeepSeek开源引擎，单卡每小时38美金立刻开赚

36氪

PHP大暴雷，全球程序员熬夜加班，超级漏洞还是ChatGPT发现的

36氪

Wed, Oct 2185

使える素材を組み合わせて広告に仕上げる

Zenn AI

Now you can chat with Gemini while using Google Maps through CarPlay

Engadget

+2 more outlets9to5GoogleAndroid Authority

Your AI agent fixed the bug, but what else did it change?

Trail of Bits

Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster

Hugging Face Blog

情シスのための Claude Code 運用レシピ集

Zenn AI

Claude Codeの自動再開のあと、古い画面に入力すると会話が二つに分かれる。送信を止めるフックを作った

Zenn AI

Deepseek: Dieses quelloffene Tool soll KI-Kosten reduzieren – für wen es gedacht ist

t3n

Responding to core PHP vulnerabilities CVE-2026-109618 and CVE-2026-109619

Google Cloud

AIエージェントに「実装しながら考える」をさせない。技術負債を減らす計画書作成プロンプト

Zenn AI

金融K線専用の基盤モデルKronosにBTC/USDTの値動きを予測させてみた

Zenn AI

Claude Code の AskUserQuestion にフックで答える（スマホからも）

Zenn AI

AI にほかの AI へ相談させていたら、返事を誰も読んでいなかった

Zenn AI

reasoning_effortが接続先で変わるときの3つの失敗と対処

Zenn LLM

ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform

OpenAI News

+1 more outletsPYMNTS

How Commerce.com and Cloudflare are getting storefronts ready for AI agents

Cloudflare Blog

Google Public Sector increases partner investments for frontline adoption

Google Cloud

OpenAI models are coming to Google Cloud

OpenAI News

+3 more outletsArs Technica AI9to5GoogleSiliconANGLE AI

How to integrate Text to Speech into your applications

ElevenLabs Blog

Talk to anyone, in any language, in realtime

ElevenLabs Blog

Google Cloud expands model choice and flexibility, bringing OpenAI models to customers

Google Cloud

Hardware-backed security at global scale: Confidential G4 with NVIDIA Blackwell is GA

Google Cloud

Google Cloud and NVIDIA expand co-engineering collaboration to scale infrastructure for the agentic era

Google Cloud

Databricks AI Runtime and NVIDIA make it easier to customize open models with enterprise data

Databricks Blog

Scaling Graph ML with NVIDIA Kumo on Snowflake ML

Snowflake Blog

AI Runtime is Generally Available

Databricks Blog

Bringing the Spark to Europe: Local Intelligence Powers Next Wave of Agents

NVIDIA Blog

NVIDIA Cloud Partners Scale Across Europe — Sovereign, Open and Built for Safe AI Agents

NVIDIA Blog

NVIDIA Advances Isaac Platform for General-Purpose Robotics With New Reference Humanoid, Frameworks, Simulation Pipelines and Data

NVIDIA Newsroom

European Mobility Leaders Build Safe Vehicles and Robotaxi Fleets on NVIDIA Hyperion

NVIDIA Newsroom

Healthcare and Life Sciences Leaders Build on NVIDIA to Advance Drug Discovery and Expand Patient Care

NVIDIA Newsroom

Tue, Oct 2092

三星8纳米再延续一代：8LPU+年内推出 主攻汽车与边缘AI芯片

cnBeta

Inception says its Mercury 3 model beats GPT-6 Luna and runs far faster

The Next Web

+2 more outletsSiliconANGLE AIThe Decoder

AMD Announces FSR 4.1 for Radeon RX 6000 Series, APUs, and Xbox

TechPowerUp

+2 more outletsWccftechTom's Hardware

Build a semantic layer for agentic AI on AWS with Denodo and Amazon Bedrock AgentCore

AWS Machine Learning Blog

LLMスクレイパーの作り方：取得・軽量化・抽出・検証の4層設計

Qiita LLM

Build a semantic layer for agentic AI on AWS with Strategy Mosaic and Amazon Bedrock AgentCore

AWS Machine Learning Blog

Build a semantic layer for agentic AI on AWS with AtScale and Amazon Bedrock AgentCore

AWS Machine Learning Blog

ALTK-Evolve: How learned memory tops the τ²-bench leaderboard on customer service tasks without a frontier model

Hugging Face Blog

「Cloudflare OS」のマネージドサービス提供へ。GitHubとの接続やGmailとの連携強化など

Publickey

Python で生成 AI の引用を検証し、回答の受理条件を判定する

Qiita LLM

DeepSeek open-sources the inference engine behind its API. And Huawei gets equal billing with Nvidia.

The New Stack

+1 more outletsTechNode

Can a local git remote replace GitHub in your agent eval?

ms-devblogs-ai

Claude Code のセッション同士でメッセージを送り合うなどスラッシュコマンドに無い機能6つ

Zenn AI

How Hamlet uses Databricks Lakebase Postgres to test changes against production data

Databricks Blog

Gemini 2.5 Flash-Liteで社内文書を整理する前に確認したいこと

Qiita LLM

LINEに送った写真と音声メモを、自宅のPCが文字にしてくれるようにした

Zenn LLM

AI同士の依頼は英語の電報調、私への報告は日本語だけ。レビューの返事は418→167トークンに

Zenn AI

Spotify pulled 50 million AI slop and spam tracks from its recommendations

The Verge AI

+3 more outletsMusic Business WorldwideThe Next WebGizmodo

Kyutai Pocket TTS v4.0.0 adds six tiny English voice models (2M–18M parameters)

Kyutai’s on-device TTS library Pocket TTS ships v4.0.0 with six new English models (18M and 7M, each with a refl variant, plus experimental 2M), selectable via --language; Kyutai calls them very strong for their size.

GitHub Releases

Linux Foundation Estimates 25% Of New Pull Requests To Be Low-Quality AI Slop

Phoronix

+1 more outletsTechPowerUp

Google DeepMind presume de haber batido a Minecraft con agentes controlados por Gemini 4 Argon en apenas 27 horas

Xataka

RSI三块关键拼图，MemT 1.0从模型层补上「记忆」

量子位

+1 more outlets量子位

AI 추론 비용 최대 10분의 1로…오픈소스 ‘llm-d 1.0’ 공개, 기업용 AI 대규모 운영 본격화

Korea AI Times

Introducing Eleven Realtime: a simpler way to build natural voice agents

ElevenLabs Blog

4,830 candidate prequalifications completed in three weeks, with data processed in Europe

ElevenLabs Blog

How to transcribe a podcast: A step-by-step guide

ElevenLabs Blog

Trinity Logistics gets 25% higher margins on shipments booked with ElevenAgents

ElevenLabs Blog

TorchTPU is now open source: Run native PyTorch on TPUs

Google Cloud

Announcing llm-d 1.0: Open distributed inference for agents on Google Cloud TPUs and GPUs

Google Cloud

AI가 게임 속 극단주의 포섭 잡는다…EU, 실시간 탐지 플랫폼 ‘워치타워’ 개발

Korea AI Times

Data updated: AI news(UTC+8)