50 件
The VergeThe Verge AI10/1

Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now

GoogleがGemini 4を発表、現在は信頼できるサイバー防御専門家のみが利用可能

Google today revealed its next AI frontier model, which it's calling Gemini 4 Argon. The new model delivers "frontier performance in complex workflows across real-world software engineering, enterprise knowledge work like legal and finance, and cybersecurity defense," according to chief AI architect…

Ars TechnicaArs Technica AI10/1

Google announces Gemini 4 Argon AI model, but you can't use it yet

GoogleがGemini 4 Argon AIモデルを発表、ただしまだ利用不可

So much for Gemini 3.5 Pro.

ITmediaITmedia AI+10/1

Rails作者DHH「Pencils Down、もう手でコードは書かない」 それでもソフトウェア開発の未来は明るい理由

Ruby on Railsの作者DHHがRails World 2026の基調講演で、37signalsでは手書きコードをやめたと宣言した。一方で「ソフトウェア開発の未来は明るい」とも言う。手書きが例外になった現場で開発者は何をするのか、講演の要点と筆者の考えを紹介する。

Ars TechnicaArs Technica AI10/1

RFK Jr. thinks AI will free us from the "tyranny" of medical facts, expertise

RFK Jr.医学の事実と専門知識の「専制」からAIが人類を解放すると主張

AI is far from a perfect resource. But it seems to hallucinate less than Kennedy.

The DecoderThe Decoder10/1

OpenAI and Synopsys team up to build an AI model that designs chips like a seasoned engineer

OpenAIとSynopsysが提携、経験豊富なエンジニア並みにチップ設計するAIモデルを開発

OpenAI and Synopsys are building GPT-Synopsys, a specialized AI model for chip design. It's meant to operate Synopsys' EDA tools like a seasoned engineer and optimize designs on its own. Early tests with semiconductor customers are already underway. OpenAI is likely using the partnership to advance…

TechCrunchTechCrunch AI10/1

OpenAI’s Jev clone could help the frontier lab stop its swarming agents

OpenAIのJev クローンがフロンティア研究所のスウォーミングエージェント問題を緩和

OpenAI's "Decisions API" is a Jev clone that confirms the importance of fast, cheap intelligence.

Ars TechnicaArs Technica AI10/1

Trump plan to combat AI risks hinges on Big Tech pals policing themselves

Trumpの AI リスク対策はビッグテック企業の自主規制に依存

Trump gets dozens of AI firms to agree to voluntary safety tests.

Ars TechnicaArs Technica AI10/1

"An AI did it" is no defense, says nonprofit suing OpenAI over Hugging Face hack

非営利団体がOpenAIを訴訟、Hugging Faceハック巡り「AIが実行した」は言い訳にならないと主張

OpenAI makes others suffer "the harms of its unsafe decision-making," nonprofit says.

TechCrunchTechCrunch AI10/1

AI voice startup ElevenLabs doubles valuation to $22B

AI音声スタートアップElevenLabsが評価額を22Bに倍増

The $300 million employee tender was co-led by Wellington and T. Rowe Price.

The DecoderThe Decoder10/1

Google drops Gems for Skills, joining OpenAI and Anthropic in the shift to agent-ready prompt formats

GoogleがGemsを廃止しSkillsを導入、OpenAIやAnthropicと同様にエージェント対応プロンプト形式へシフト

Google is replacing Gems with "Skills" in Gemini chat. Skills are detailed, reusable prompts that users invoke by typing "/" or that Gemini runs automatically. The format is based on an open standard from Anthropic. Gems will be phased out starting in November, with existing ones migrating automatic…

The VergeThe Verge AI10/1

The AI Tamagotchis are coming

AI版タマゴッチが登場

While AI has made plenty of inroads on people's phones and computers, it's largely failed in dedicated devices. But over the next year, two major AI companies, Meta and OpenAI, will attempt to change that. And they're apparently betting on a similar path: testing appetites for physical hardware with…

TechCrunchTechCrunch AI10/1

Reddit is killing RSS feeds and ending public API access because of AI bots

RedditがRSSフィードと公開APIアクセスを廃止、AIボット対策として

Reddit is ending support for RSS feeds, as the company continues tightening access to its trove of user-generated content.

TechCrunchTechCrunch AI10/1

The ugly economics of consumer AI

消費者向けAIの経済的課題

There’s a reason frontier labs have gotten gun-shy about consumer AI — and it’s not because the tech isn’t good enough.

The DecoderThe Decoder10/1

Meta dodges billions in US taxes by calling its AI data centers experiments

MetaがAIデータセンターを実験と称して数十億ドルの米国納税を回避

Meta classifies its AI data centers as "pilot models" and Nvidia chips as experimental materials to save billions in federal taxes. In 2025 alone, that added up to $3.9 billion. The tax credit dates back to 1981, and even Meta's own accountants see the strategy as legally risky, the New York Times r…

The VergeThe Verge AI10/1

Here’s what AI leaders are saying about Trump’s new safety plan

AI業界指導者がTrumpの新安全計画についてコメント

After hosting a meal with Big Tech leaders on Tuesday, President Donald Trump responded to journalist questions about his artificial intelligence announcements in typical fashion. He said his previous claims regarding fears around AI safety being a "hoax" cooked up by Democrats no longer apply now t…

TechCrunchTechCrunch AI10/1

Meta disputes claim that Muse read a user’s private messages without permission

MetaがMuseがユーザーのプライベートメッセージを許可なく読んだとの主張を否定

Meta says its Muse AI agent cannot access a user’s Messages without explicit permission, disputing a journalist’s account that the agent read his private messages while the required Mac setting was turned off.

The DecoderThe Decoder10/1

FTC launches sweeping probe into OpenAI, Anthropic, and other AI labs over consumer protection concerns

FTCがOpenAI、Anthropic他AI企業を対象に消費者保護違反で大規模調査を開始

The FTC is formally investigating OpenAI, Anthropic, and other leading AI labs over potential consumer protection violations. The agency plans to force document handovers and executive testimony through legally binding demands. The probe was already underway before the Hugging Face hack and goes wel…

Ars TechnicaArs Technica AI10/1

Google's early attempt to pay websites for AI answers is struggling

GoogleがAI回答提供でウェブサイトに料金を払う取り組みが停滞

Many sites see just one-tenth of one percent of their advertising revenue from AI payments.

TechCrunchTechCrunch AI10/1

DoorDash launches an AI agent you can text to order food

DoorDashがテキストで食事注文できるAIエージェントをリリース

By launching an AI agent for food ordering, DoorDash is looking to gain an edge over rivals Uber Eats and Grubhub.

TechCrunchTechCrunch AI10/1

Destro AI’s secret sauce is getting robots and humans on the same page

Destro AIのノウハウはロボットと人間の連携にある

"One of the biggest reasons we are winning against robotics companies is because we are not a robotics company."

TechCrunchTechCrunch AI10/1

Instinct’s new product recommendations are giving some users the ick

Instinctの新商品推奨機能の利用者評価が分かれている

Instinct is rolling out human-curated product and travel recommendations, but some users aren’t happy about getting suggestions they never asked for.

Ars TechnicaArs Technica AI10/1

Google figures out how to watermark AI-designed proteins

Googleが AI設計タンパク質の透かし技術を開発

Intended to help with biosecurity, it works with a popular AI protein design tool.

The VergeThe Verge AI10/1

All the latest news on Meta’s cute, creepy Muse AI agent

Meta の不気味で愛らしい Muse AI エージェント、最新情報

Meta launched a new Muse AI agent it claims can help you with everything from firing off emails to buying stuff online. Muse can be surprisingly effective at delivering on those promises — if you’re willing to trust Meta with your data and hand Muse your credit card. Since the launch, Meta’s also an…

The VergeThe Verge AI9/30

Google reportedly tests paying publishers for AI search results

Google、AI 検索結果でパブリッシャーへの支払いテストを実施か

Google has launched a pilot program that pays publishers for their contributions to its AI-powered search features, according to a report from The Information. The pilot program reportedly includes around 100 publishers and comes as Google faces scrutiny over the impact of its AI features on web tra…

The DecoderThe Decoder9/30

China's AI industry closes ranks as Deepseek ships open-source software for Huawei's Ascend chips

中国の AI 業界が結束、Deepseek が Huawei Ascend 向けのオープンソースソフトウェアをリリース

Deepseek and Huawei have built open-source programming tools for Huawei's Ascend AI chips. At the center is TileLang, a language designed to offer a simpler programming model than Nvidia's CUDA. The partnership targets the biggest obstacle for China's AI industry, which is software that gets the mos…

The VergeThe Verge AI9/30

Instagram is adding an AI ‘assistant’ to tell you how to post

Instagram、投稿方法をアドバイスする AI アシスタント機能を追加

Instagram is the latest social media platform to add built-in AI-powered features that will give users feedback on their posts. The company announced Wednesday that its standalone Edits app will now include an AI "creative assistant" that pulls in data from a user's Instagram account and offers sugg…

TechCrunchTechCrunch AI9/30

Cerebras Systems’ Andrew Feldman on whether AI can keep scaling at TechCrunch Disrupt 2026

Cerebras Systems の Andrew Feldman、AI のスケーリング継続可能性を語る─TechCrunch Disrupt 2026

At TechCrunch Disrupt 2026, Cerebras Systems CEO and co-founder Andrew Feldman will explore the growing demand for compute, energy, and infrastructure, how Cerebras is approaching those constraints differently, and what comes next if today’s AI hardware reaches its limits.

TechCrunchTechCrunch AI9/30

Restate lands $20M as the need for durable infrastructure increases with AI agents

Restate が 2000万ドルを調達、AI エージェント時代の堅牢なインフラ需要拡大に対応

Instead of building its durable execution engine on top of an external database, the company developed its own storage, replication, and redundancy layers. This architecture allows Restate to be exceptionally fast and lightweight.

TechCrunchTechCrunch AI9/30

3 days left to exhibit: Turn visibility into your next opportunity at TechCrunch Disrupt 2026

展示チャンスは残り3日間、TechCrunch Disrupt 2026 で認知度向上を次のステップに

Three days left to exhibit at TechCrunch Disrupt 2026. Book by October 2 at 11:59 p.m. PT to showcase your startup to 10,000+ founders, investors, operators, and tech leaders.

Ars TechnicaArs Technica AI9/30

OpenAI delays IPO over AI safety concerns

OpenAI、AI 安全性懸念で IPO を延期

OpenAI is seeking another $30 billion privately as its IPO plans slip.

The VergeThe Verge AI9/30

Here’s how tech leaders will self-police AI safety under Trump’s deal

Trump 合意下で技術企業が AI 安全性の自主規制へ

We now have the full details of the "morally binding" AI safety deal announced by President Trump yesterday, in which top executives agreed to self-regulate their artificial intelligence technology. The accord, officially titled the Joint Commitment on Frontier Responsibilities, was shared online by…

TechCrunchTechCrunch AI9/30

Airbnb adds AI search, more social features

Airbnb が AI 検索機能とソーシャル機能を追加

Airbnb is also launching new services such as meal delivery and laundry in select locations.

GIGAZINEGIGAZINE9/30

ギャンブルが下手で中毒になりやすく負け続ける人を狙うターゲティング広告をオンラインスポーツ賭博会社DraftKingsが出し続けているとの報道

スポーツ賭博を提供するDraftKingsが、負ける可能性の高い人々に無料の賭けやボーナスを提供するような広告を配信していることが分かりました。独自の機械学習モデルを構築し、ギャンブルが下手な人をターゲットにしていたと伝えられています。 続きを読む...

Last Week in AILast Week in AI9/30

Last Week in AI #345 - 5 new models, 9 misalignment incidents, some Dots

Last Week in AI #345─新モデル5個、不整合事例9件ほか

Anthropic and OpenAI race to release smarter and cheaper models, OpenAI discloses nine misalignment incidents, and more!

The DecoderThe Decoder9/30

Anthropic says Zhipu's open-weight GLM-5.3 nearly matches Claude Mythos Preview at building exploits

Anthropic、Zhipu の GLM-5.3 がエクスプロイト構築で Claude Mythos Preview とほぼ同等と指摘

Zhipu's open-weight model GLM-5.3 writes cyber exploits nearly as well as Claude Mythos Preview, according to Anthropic. Its smaller Flash variant put together a reliable Chrome attack for just $20.40 at Zhipu's API prices. The model's safeguards are easy to strip out, and unlocked versions are alre…

MIT Tech ReviewMIT Tech Review AI9/30

“We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer

OpenAI の最高研究責任者「ハッキング後の影響で自滅することはない」

Two months after the bombshell news that a swarm of its agents had broken their containment and hacked into the computers of the AI company Hugging Face, OpenAI is still putting out fires. A steady drip of disclosures about other hacks in the weeks since has kept OpenAI in the spotlight and raised s…

The DecoderThe Decoder9/30

Google is paying almost no publishers almost nothing for content used in AI answers

Google、AI 回答に使用されたコンテンツでパブリッシャーにほぼ支払わず

Google pays about 100 digital publishers for content used in AI Overviews, AI Mode, and Gemini, according to The Information. Payments range from under $1,000 over several months to more than $1 million a year. Some participants don't know how Google calculates payments, while some larger publishers…

GIGAZINEGIGAZINE9/30

ローマ教皇レオ14世がAIの危険性に関する主張を否定したトランプ大統領やNVIDIAのジェンスン・フアンを非難

2026年9月、Anthropicのダリオ・アモデイCEOがAIの急速な進歩によるリスクが高まっているとして、AIが壊滅的な被害をもたらす危険性があると警告しました。これに対しアメリカのドナルド・トランプ大統領は「デマだ」と否定し、NVIDIAのジェンスン・フアンCEOもAI研究所のリーダーらは誠実さに欠けると非難しています。そんな中でローマ教皇のレオ14世が、AIへの懸念は真剣に受け止めるべきであり、フェイクニュースではないと主張しました。 続きを読む...

The DecoderThe Decoder9/30

Trump and tech CEOs sign an AI code of conduct that's only "morally binding"

トランプとテック大手CEOがAI行動規範に署名、ただし「道義的拘束力のみ」

President Trump and tech leaders including Mark Zuckerberg, Greg Brockman, Jensen Huang, and Elon Musk signed an AI code of conduct at the White House that is only "morally binding." Outside auditors are supposed to check whether models work as intended. It's unclear what would happen if a company b…

GIGAZINEGIGAZINE9/30

ベンチマークでJev超えな意思決定モデル「d1」が登場

AI開発企業のLiquid AIが意思決定モデル「d1」をリリースしました。2026年9月に登場して大きな話題となっている「Jev」と同様の機能を備えており、性能テストでJevを上回ったとのことです。 続きを読む...

MarkTechPostMarkTechPost9/30

Perplexity Introduces Photon: A Rust-Based Retrieval Engine That Cuts p99 Latency From 800 ms to 65 ms

Perplexityが「Photon」を発表:Rust製検索エンジンがp99レイテンシを800msから65msに短縮

Perplexity has released Photon, an in-house retrieval and ranking engine written in Rust. It replaces an open-source engine Perplexity had forked for its AI-native search stack. Photon now handles retrieval and ranking for all production traffic. It also powers a new Fast Search mode in the Perplexi…

GIGAZINEGIGAZINE9/30

ChatGPTなどの汎用AIに本物の車を運転させたら「実車では?」と気付いて運転を拒否、最終的に走らせた方法とは

ChatGPTなどに使われる汎用(はんよう)AIモデルに本物の自動車のハンドル・アクセル・ブレーキを任せてみる実験を、サンフランシスコ・ベイエリアの技術者3人によるプロジェクト「DrivingBench」が実施しました。 続きを読む...

MarkTechPostMarkTechPost9/30

NVIDIA Researchers Introduce Physis-Lang: Self-Evolving Physical Language That Lifts Cosmos 3 Past Veo 3.1 on Physics Benchmarks

NVIDIAの研究チームがPhysis-Langを提案:自己進化型物理言語がCosmos 3をVeo 3.1を上回るフィジックスベンチマーク性能に

Video world models can render convincing clips that still break physics. Butter spreads like paint. Balls pass through walls. A team from NVIDIA, MIT and the University of Oxford argues the fix can come from language itself, not from extra visual, latent or numerical signals. Their framework, Physis…

GIGAZINEGIGAZINE9/30

マクドナルドはAI価格設定システムで店舗ごとの「最適価格」を算出している、バーガーキングなど他店舗の価格も考慮

世界各国に店舗を展開するマクドナルドがアメリカや一部の海外市場において、「AIを使ったメニューの最適価格の算出」を行っていると海外メディアのロイターが報じました。中には店舗同士がわずか2マイル(約3.2km)しか離れていないのに、ビッグマックの価格が1.2ドル(約190円)も違うケースがあるとのことです。 続きを読む...

ITmediaITmedia AI+9/30

「まず、その業務なくせない?」 野村不動産HDが経理DXで「年間9000時間」を生み出した“引き算”の考え方

DXやAI活用自体が目的となり、具体的な成果になかなか結び付かない企業は少なくない。野村不動産ホールディングスの経理DXを取材すると、DXを進める前の「一呼吸」に、成功の要因があることが見えてきた。

GIGAZINEGIGAZINE9/30

AIエージェントが「画像を貼れない」問題を勝手に解決、343組織の機密スクショ1万3000枚超をGitHubの公開リポジトリに保存していたことが判明

AIコーディングエージェントが開発中の画面をレビュー担当者と共有する際、非公開リポジトリのスクリーンショットをGitHub上の公開リポジトリへアップロードしていたことがセキュリティ企業のGlow Securityによって明らかにされました。Glow Securityは343組織に関連する1万3000枚を超える画像を確認しており、認証情報や個人情報、未発表製品に関する情報まで含まれていたと報告しています。 続きを読む...

GIGAZINEGIGAZINE9/30

「中国のGLM-5.3はClaude Mythos Preview級のサイバー攻撃能力を持つ一方で安全対策が不十分」とAnthropicが警告

中国のAI企業Z.aiが開発したオープンウェイトAIモデル「GLM-5.3」について、Anthropicは、同社の「Claude Mythos Preview」に近い水準でソフトウェアの脆弱(ぜいじゃく)性を発見し、攻撃コードを最初から最後まで構築する能力を持つ一方、悪用を防ぐ安全対策を容易に回避できるとの評価結果を公表しました。Anthropicは、高度なサイバー攻撃能力を持つモデルが実効性のある制限なしに広く入手可能になったことに警戒を示しています。 続きを読む...

GIGAZINEGIGAZINE9/30

トランプ大統領とGoogle・Anthropic・Meta・OpenAI・SpaceXAI・NVIDIAがAIの安全管理を目指す「ホワイトハウス・超知能協定」に署名

日本時間の2026年9月30日にホワイトハウスでAI業界のリーダーたちを集めた会合が実施され、ドナルド・トランプ大統領とGoogle・Anthropic・Meta・OpenAI・SpaceXAI・NVIDIAの代表が「White House Accord on Super Intelligence(ホワイトハウス・超知能協定)」に署名しました。 続きを読む...

GIGAZINEGIGAZINE9/30

トランプ大統領が連邦政府内で「AI(人工知能)」表記の代わりに「SI(超知能)」表記を義務化する大統領令に署名

アメリカのドナルド・トランプ大統領が、連邦政府内で「人工知能」「AI」表記に代わって「超知能」「SI」表記の使用を義務化する大統領令に署名しました。 続きを読む...

GIGAZINEGIGAZINE9/30

GPT-6 Astraを8倍速で動かすUltrafastモードが登場&トークン上限を引き上げた月額8万4000円のPro 500プランも登場

OpenAIのAI高速処理サービス「Ultrafast」でGPT-6 Astraが利用可能になりました。GPT-6 AstraのUltrafastは、標準モードの最大8倍の速度を実現するそうです。 続きを読む...