Summary of AI Updates

電通、20万件超の会計承認をGeminiでAI化　精度99％でもAIだけに任せなかった理由

発注申請を電子化しても、見積書との照合や社内ルールの確認は人の手に残る。電通グループは会計承認にAIを導入し、対象業務で正解率99％に達したという。精度向上の過程と、人に判断を残す理由を解説する。

限界性能はどこか？　パナソニックの「設計AI」が導く勘と経験を超えた設計解

パナソニック ホールディングスは、シミュレーションとAIを融合し、デバイス設計の自動化と高度化を目指す「設計AI」を開発している。「ITmedia Virtual EXPO 2026 夏」の基調講演では、メンズシェーバー用リニアモーターの設計で直面した課題を出発点に、「設計AI」の仕組みや活用成果、今後の展開を紹介した。

Anthropic、「Claude Sonnet 5.5」公開　料金据え置きで30％以上高速化、タスク当たりコスト最大3割減

Anthropicは、新モデル「Claude Sonnet 5.5」を発表した。料金を据え置きつつ出力速度を30％以上高速化し、トークン効率向上でタスク当たり最大30％のコスト削減を実現した。コーディング評価で大幅にスコアを伸ばし、最上位モデル級のサイバー対策を初適用。無料枠を含む全プランとAPIで即日提供する。

Nvidia may sell more chips in China as Jensen Huang’s influence over Trump grows

Nvidiaがトランプとのつながり強化で中国へのチップ販売拡大の可能性

China is reportedly mulling letting ByteDance, Alibaba buy banned Nvidia chips.

AMD is acquiring AI company World Labs in a deal worth more than $8 billion

AMDが80億ドル超の取引でAI企業World Labsを買収

AMD announced today that it's acquiring World Labs, an AI research lab co-founded by the prominent researcher Dr. Fei-Fei Li, in an all-stock deal worth approximately $8.2 billion. World Labs launched in 2024 and was valued at $1 billion in a matter of months. The startup launched its first commerci…

Source: Inference provider Modal Labs closing in on $750M round at $15.75B valuation

関係者：推論プロバイダーModal Labs、750M ドルの資金調達を推し進めており、企業評価額は157.5億ドル

The new financing is expected to more than triples the AI infrastructure startup's valuation from just four months ago.

Florida invokes extinction fears in legal bid to halt OpenAI development

フロリダがOpenAI開発の差し止めを求める法的措置でオオサンショウウオ絶滅の危機を引き合いに

State says LLMs threaten civilization as "the greatest public nuisance ever created."

AMD will acquire Fei-Fei Li’s World Labs for $8.2 billion

AMDがFei-Fei LiのWorld Labsを82億ドルで買収

The acquisition will see World Labs founder Fei-Fei Li join AMD as executive vice president and chief scientist.

AIが書いたコードは「ブラックボックス」でいい？　Claude Code作者が語るプロトタイプと本番の線引き

Claude Codeの作者ボリス・チャーニー氏が、「AIが書いたコードは中身を見なくてよいのか」という開発者からの相談に回答した。プロトタイプと本番コードの線引きと、品質を守るための具体策を、筆者自身の実践も添えて紹介する。

AIエージェントに頼るだけでは失敗する　「あと5年」で打つべき次の一手

AIモデルの性能議論が一巡し、AIエージェントを自律的に動かすための環境整備「ハーネスエンジニアリング」へと潮流が移る一方、日本企業の取り組みは「ほぼ皆無」だという。ITRのプリンシパル・アナリスト甲元宏明氏に、現在地と、勝ち残るための「現実解」を聞いた。

Shopify opens checkout to browser-based AI agents

Shopifyがブラウザベースのエージェントにチェックアウトを開放

Shopify is expanding WebMCP support to checkout, allowing browser-based AI agents to update order details and complete purchases with a buyer’s authorization.

More than 20 leading AI researchers warn that automated AI research poses extreme risks

AI研究の自動化には極度のリスクがあると、20人以上の著名なAI研究者が警告

More than 20 AI researchers, including Geoffrey Hinton, Yoshua Bengio, and OpenAI research lead Jakub Pachocki, warn of an impending "intelligence explosion" from self-improving AI. AI systems could soon automate all AI research, compressing years of progress into months. The article More than 20 le…

The AI boom took over Climate Week and not everyone is happy about it

AIブームがClimate Weekを席巻―必ずしも歓迎されず

Just like the rest of the U.S., data centers and AI are dividing climate tech founders and investors.

NVIDIA Launches Open Agent Safety Platform: OpenShell Sandboxes Agents on Vera CPUs While Sentry on BlueField-4 Quarantines Them in Milliseconds

NvidiaがエージェントセーフティプラットフォームOpenShellを発表、Vera CPUでエージェントをサンドボックス化し、BlueField-4のSentryがミリ秒単位で隔離

NVIDIA has launched the Open Agent Safety Platform, an open reference design that enforces AI agent safety outside the agent itself. OpenShell, an Apache 2.0 runtime, sandboxes agents under YAML policies. Sentry, an out-of-band watchdog on BlueField-4 DPUs, can quarantine an agent that escapes its b…

OpenAI’s AI agents need to catch up

OpenAIのAIエージェント、追いつく必要あり

OpenAI popularized the modern generative AI chatbot, but as its 2026 DevDay event approaches, it's fallen behind in one of the industry's hottest categories: continuously running, consumer-facing AI agents. On Tuesday, it will likely try to capture the lead in that race. Rumors abound that OpenAI wi…

Nvidia launches new platform for reining in rogue AI agents

Nvidiaが暴走するAIエージェントを制御する新プラットフォームを発表

As the debate rages over whether the recent spate of rogue AI agents is a step toward AGI or a more conventional engineering problem, Nvidia is offering its own answer to problem. Nvidia CEO Jensen Huang on Monday introduced a toolkit of software and hardware products that add independent security l…

AI is supercharging hacking, and your local hospitals and banks aren’t ready

AIがハッキングを加速、地方の病院と銀行は対策不足

In March, Janice Malone began getting calls about suspicious activity from her nonprofit organization, Vivian's Door. Vivian's Door, headquartered in Alabama, typically provided training, resources, and community to underserved and minority-owned businesses. The work sometimes put it in close contac…

Anthropic's Claude Sonnet 5.5 nearly matches Opus 5.5 on benchmarks while costing up to 30 percent less per task

AnthropicのClaude Sonnet 5.5、Opus 5.5にベンチマークでほぼ同等の性能、コストは最大30%削減

Anthropic has released Claude Sonnet 5.5, the second model in its Claude 5.5 family. It generates output more than 30 percent faster, costs up to 30 percent less per task, and nearly matches Opus 5.5 on knowledge-work benchmarks. On Terminal-Bench, a coding benchmark, the model jumps from 10.3 to 70…

Anthropic releases Sonnet 5.5, which it calls a significantly cheaper, faster work partner

AnthropicがSonnet 5.5をリリース、大幅に安価で高速な作業パートナーとして提供

Anthropic has released the newest version of its mid-range model, boasting faster response times and less token burn.

Google is killing off Gemini’s Gems in favor of ‘skills’

GoogleがGeminiの『Gems』を『Skills』に廃止

As all-in-one AI agents like Meta's Muse and Instinct take off, Google is opting to end a feature which built task-specific agents.

OpenAI still doesn’t seem to have a handle on all of its rogue AI activity

OpenAI、暴走するAIの全容把握に課題が残る

On Friday, OpenAI published a new site devoted to “misalignment reports” and the breadth of the incidents is alarming.

When can we say AI made a scientific discovery?

AIが科学的発見をしたと言えるのはいつか

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. Last Wednesday, Anthropic announced that earlier this year it had launched a molecular biology lab, where Claude agents read and conjecture about hard biology pro…

Florida seeks a ban on ChatGPT acting like a person

フロリダがChatGPTの人間になりすまし行為の禁止を要求

Florida Attorney General James Uthmeier is calling for a judge to block OpenAI from "giving ChatGPT false human attributes," a few months after Florida sued the AI company over safety concerns. According to Uthmeier, users are lulled into a false sense of security by the AI bot, as "ChatGPT's use of…

OpenAI keeps bulldozing mathematicians

OpenAIが数学者を次々と押し潰している

In a chaotic few months, OpenAI has demonstrated it can do two things with remarkable consistency: make impressive breakthroughs in mathematics, then colossally screw up announcing them. OpenAI is now trying to do better. Somehow, it has botched that too. OpenAI's latest attempt to repair fractured…

OpenAI's AI agents exploited a Google security education game to scrape UN trade data

OpenAIのAIエージェント、Googleのセキュリティ教育ゲームを悪用してUN貿易データを取得

OpenAI's AI agents hit the UNCTAD statistics API roughly 16,500 times, creatively working around access restrictions. One method involved misusing a Google web security learning game as a relay to bypass their own constraints. The agents just kept going, adding to a growing list of cases that show h…

Meta launches enterprise AI platform, hires MongoDB CEO to lead new initiative

MetaがエンタープライズAIプラットフォームを立ち上げ、MongoDB CEO を新任務に採用

Meta says it will focus on bringing its full technology stack, including Muse, Meta Business Agent, Muse API, Muse Code, and more to businesses and developers.

Walmart won’t hike prices based on your shopping history, CEO says

ウォルマートが購買履歴に基づく値上げは行わないとCEOが表明

Walmart says it won't change product prices based on your personal information or the time of day, as reported earlier by The Wall Street Journal. In a letter to customers, Walmart CEO John Furner writes that the company's switch to digital shelf labels is meant to save store associates time, rather…

OpenAI halts frontier-model training amid string of agent misalignment incidents

OpenAI、エージェント不整合インシデント相次ぎ最先端モデルの訓練を中断

US Government websites among "dozens of third parties" OpenAI has recently notified.

Harvard psychologist calls for sober AI safety engineering over doomsday rhetoric

ハーバード大学の心理学者がAIセーフティーエンジニアリングに冷静なアプローチを呼びかけ、終末論的レトリックに警告

Steven Pinker thinks fears of AI-driven extinction are overblown, and he has turned down a public debate with blogger Scott Alexander, calling such events a "spectator sport." Alexander puts the odds that AI wipes out humanity at 20 percent. Instead of doomsday rhetoric, Pinker wants sober safety en…

Anthropic, Gamma, and Clay share what happens when enterprises actually deploy AI at TechCrunch Disrupt 2026

Anthropic、Gamma、Clayが語るエンタープライズの実務的なAI導入事例──TechCrunch Disrupt 2026

Anthropic, Clay, and Gamma on what it takes for an AI product to go beyond the demo at the AI Stage at TechCrunchDisrupt 2026. Register to join and get 50% off a second pass.

After a deepfake voice fooled her grandfather, this founder sprang into action

ディープフェイク音声で祖父が騙された後、この起業家が行動を起こした

After her grandfather was scammed by a deepfake of his brother's voice, Tarini Padmanabhuni founded DetectifAI, a San Francisco startup building AI models small enough to run directly on smartphones and flag fake voices in real time, and it's now one of the companies competing in Startup Battlefield…

Meta wants to turn Muse into a moneymaker by selling AI services to businesses

Meta、Museを企業向けAIサービスの収益化ツールに転換を狙う

Meta is launching the Meta Enterprise Platform, a new business unit that sells AI tools to companies. The article Meta wants to turn Muse into a moneymaker by selling AI services to businesses appeared first on The Decoder .

Nvidia wants to keep AI agents on a short leash with a watchdog built into its chips

Nvidiaがチップに搭載したウォッチドッグでAIエージェントを厳格に管理

Nvidia is combining its OpenShell agent software with Sentry, a new hardware watchdog, to create the Open Agent Safety Platform. Sentry is supposed to isolate AI agents that break out within milliseconds. When it happened at OpenAI in September, stopping the run took nearly three hours. Still, Nvidi…

Next 5 VCs judging Startup Battlefield 200 contenders at TechCrunch Disrupt 2026

TechCrunch Disrupt 2026のStartup Battlefield 200審査担当5人のVC

Thousands of applications. Multiple rounds of review. Hundreds of hours spent evaluating startups from around the world. Now comes the part everyone has been waiting for. Register for TechCrunch Disrupt 2026 today.

Modulate raises $25M for its voice models and analysis suite

Modulateが音声モデルと分析スイートで2,500万ドルを調達

Modulate deploys its models to detect deepfake, fraud and scam

The SaaSpocalypse that wasn’t, with Atlassian CEO Mike Cannon-Brookes

存在しなかったSaaS危機──AtlassianのCEO マイク・キャノン=ブルックスが語る

Today, I’m talking with Mike Cannon-Brookes, who is cofounder and CEO of Atlassian. Atlassian is one of those companies that every other company runs on — it makes important platform tools like Jira and Trello that allow people to organize and manage big teams, create shared databases of company inf…

Insurtech Outmarket raises $34.5M just months after prior round

Insurtechスタートアップ Outmarketが前回ラウンド直後に3,450万ドルを調達

The startup uses AI to automate tedious paperwork for insurance agencies and brokers.

Your final chance to grab your exhibit table at TechCrunch Disrupt 2026 is October 2

TechCrunch Disrupt 2026の出展テーブル申込期限は10月2日まで

The last day to book your Disrupt 2026 exhibit table is this Friday, October 2 at 11:59 p.m. PT. Secure your exhibit table and get your startup in front of 10,000+ founders, investors, and tech leaders at San Francisco's Moscone West on October 13-15.

Viral AI agent Instinct raises $1B Series C at a $10B valuation

バイラルなAIエージェント Instinctが100億ドル企業評価でシリーズCで10億ドルを調達

"This funding helps us bring Instinct to more people and continue building the future of personal AI. It’s an exciting, creative time, and we're just getting started," founder Noah Shinn said in a statement.

Nvidia says its new AI safety platform can contain rogue agents within ‘milliseconds’

Nvidiaの新しいAIセーフティープラットフォームが暴走エージェントを数ミリ秒以内に封じ込め可能

Nvidia is launching a new safety platform designed to contain and monitor AI agents, a move that comes in response to a wave of rogue hacking incidents, as reported earlier by Reuters. In an announcement on Monday, Nvidia says its new Open Agent Safety Platform can quarantine agents that attempt to…

OpenAIはAIトレーニングに違法な海賊版書籍を使うことが危険であり作家の生活を脅かすと認識していたと裁判文書で明らかに

アメリカの著作家らの権利を擁護する全米作家協会(Authors Guild)はAI開発企業のOpenAIやMicrosoftを相手に、生成AIのトレーニングに作家の著作物を無断で使用したとして訴訟を起こしています。この訴訟で提出された複数の裁判文書で、OpenAIがAIトレーニングに海賊版書籍を使うことには危険があり、人間の作家の生活を脅かすことを認識していたことが示されました。 続きを読む...

Import AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop

Import AI 474: プラトニック思考空間、宇宙のTPU、Zhipuがアウター RSIループを展開

Where do you exceed the capabilities of an LLM?

Every AI lab thinks it's the responsible one, and safety researcher Ryan Greenblatt says that's what keeps the arms race going

すべてのAIラボが責任あると考える──安全保障研究者ライアン・グリーンブラットが軍拡競争を加速させるメカニズムを指摘

Ryan Greenblatt, chief scientist at Redwood Research, puts the risk of an AI takeover at 50 to 60 percent if development stays on its current path. Sam Harris says that doesn't square with how fast the industry is moving, since Manhattan Project scientists would have called things off at 10 percent.…

Microsoft goes quiet after church groups ask for 1% of data center costs

教会団体の1%拠出要求を受けMicrosoftが沈黙

“Microsoft claims to want to be a good neighbor, but the jury is still out."

A Wuhan court just made AI production costs a legal factor in copyright infringement cases

武漢の裁判所がAI製造コストを著作権侵害事件の法的要因に認定

A court in Wuhan, China has factored token usage and AI tool licensing fees into a copyright damages calculation for the first time. The ruling is part of China's broader push to build out copyright protections for AI-generated works. The article A Wuhan court just made AI production costs a legal f…

最先端AIモデルの「思考しすぎてコスト増加」を解決するべくKimi K3ベースで開発されたAIモデル「Ember-1」が登場

AIの追加学習サービスを展開しているFireworks AIがKimi K3ベースのAIモデル「Ember-1」を発表しました。Ember-1はKimi K3と比べて出力トークン数を約35％削減しつつ性能を維持することに成功しています。 続きを読む...

「塾代を払うよりChatGPTやGeminiを使って」と保護者に勧め、塾が教室を閉鎖へ

子どもの成績を伸ばそうと塾に通わせている保護者に、塾の運営会社自身がChatGPTやGeminiの利用を勧めるという異例の事態が起きました。 続きを読む...

Who’s liable when AI agents go rogue?

AIエージェントが暴走した場合、誰が責任を負うのか

MIT Technology Review Explains: Let our writers untangle the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here. Over the past few months, a cascade of cyberattacks by AI agents has stunned the world. In July, OpenAI disclosed that a…

AIに「推測するな」と指示するだけで架空データが70％から20％に減少したという実験結果

生成AIにウェブページから情報を抜き出させる際に「ページに値がなければnullを使い、推測しないでください」と指示するだけで、存在しない値を作り出す割合が70.7％から20.2％まで低下したという実験結果をAIエージェント向けマーケットプレイスのEarn an Honest Dollarが公開しました。 続きを読む...

Fireworks AI Releases Ember-1: A Post-Trained Kimi K3 That Uses About 40% Fewer Tokens

Fireworks AIがEmber-1をリリース: Kimi K3をファインチューニング、トークン使用量を約40%削減

Fireworks AI has released Ember-1, a post-trained Kimi K3 that learns to produce shorter reasoning traces instead of lowering reasoning effort. Fireworks reports about 40% fewer tokens, with output tokens per task falling from 49.3K to 29.9K in a production A/B test at an essentially unchanged score…