The companies that spent July being regulated pushed back — though the reporting disagrees about who joined. Business Insider has Microsoft, Meta, Nvidia, OpenAI, and Palantir signing an open letter urging US policymakers not to regulate open-weight AI models, and TechRadar notes OpenAI signed as the White House accuses Moonshot AI of stealing from Anthropic. A later account describes a 25-firm coalition letter that OpenAI and Anthropic both skipped, calling the absence a lobbying split. The accounts may describe different letters; treat the signatory list as unsettled. What is not in dispute is the ask: no premature restrictions on open weights.
The model shelf refilled. OpenAI launched the GPT-5.6 family — Sol, Terra, and Luna — alongside ChatGPT Work, an agent pitched to manage apps, files, and long-running projects on its own. Anthropic’s Claude Opus 5 drew strong benchmark notices at competitive pricing. And Google’s absence kept compounding: one tracker puts Gemini 3.5 Pro at 67 days past its promised release while rivals ship.
Korea spent the day assembling an AI buildout. NAVER, NVIDIA, and Brookfield will expand a national AI factory to 200 megawatts with gigawatt ambitions — a $1 billion NVIDIA investment and up to $9 billion from Brookfield — while President Lee met Nvidia, OpenAI, Anthropic, and Broadcom CEOs in San Francisco, a meeting one report says included a $500 billion Nvidia–SK Group partnership (a single-source figure worth reading carefully). Regulators moved in parallel: the Digital Omnibus is approved, and not everything is delayed, Delhi’s High Court warned an injunction against OpenAI could hurt Indian AI development, and Brazil’s AI election-ad rules took effect. Meanwhile the jobs panic had a skeptic day — the Guardian argues the AI jobs apocalypse probably isn’t coming anytime soon — even as CNBC describes a coalition of religious leaders, unions, and local activists forming against AI’s societal costs.
What People Are Asking
Is Claude Opus 5 out?
Weekly coverage now treats it as released — Patrick McGuinness's AI Week in Review covers the Claude Opus 5 release alongside Poolside's Laguna S 2.1 and Gemini 3.6 Flash, with benchmark comparisons. AI Week in Review →
What is the open-weight AI letter?
An open letter signed by Microsoft, Meta, Nvidia, OpenAI, and Palantir urging US policymakers not to impose premature restrictions on open-weight AI models. Business Insider →
Is the White House accusing China of stealing AI?
Per one report, yes — the White House accuses Chinese firm Moonshot AI of intellectual-property theft from Anthropic, in the same news cycle as the industry's open-weight letter. TechRadar →
Is the EU AI Act delayed?
Partly — the approved Digital Omnibus postpones high-risk obligations, but transparency and AI-literacy requirements still apply from August 2026. Anybody Can Prompt →
What is Korea's AI factory deal?
NAVER, NVIDIA, and Brookfield plan to expand Korea's AI factory infrastructure to 200 megawatts with gigawatt-scale potential — including a $1 billion NVIDIA investment and up to $9 billion from Brookfield. GlobeNewswire →
Did Nvidia really commit $500 billion to SK Group?
One report of President Lee's San Francisco meetings says Nvidia announced a $500 billion SK Group partnership — a single-source figure worth reading skeptically until corroborated. Seoul Economic Daily →
Is AI actually taking jobs yet?
Recent studies suggest not at scale — the Guardian's Eduardo Porter argues significant displacement hasn't materialized, and a separate analysis finds the impact concentrated at entry level, with companies overstating AI as the cause of layoffs. The Guardian →
Why was the robot teacher plan paused?
Salamanca's school district halted its AI humanoid robot-teacher plan after New York state officials raised student-data privacy concerns. Utica Phoenix →
What is GPT-5.6?
OpenAI's newly launched model family — Sol, Terra, and Luna — shipped alongside ChatGPT Work, an agent pitched to manage apps, files, and long-running projects autonomously. The Agency Journal →
Every Story From July 25
Jump to: AI infrastructure (17) · AI labs and models (13) · Policy and regulation (7) · Society, politics, and ethics (5) · Consumer tech and culture (2) · Labor and workplace (2) · Markets and macro (2) · Security and agent risk (2)
AI infrastructure17 stories
Digital Applied's analysis of OpenAI's $30B Georgia campus highlights persistent compute scarcity and rising costs through 2028, advising buyers to plan for rationing.
NAVER, NVIDIA, and Brookfield announce plans to expand Korea's AI factory infrastructure to 200 megawatts with potential for gigawatt-scale deployment, including a $1 billion NVIDIA investment and up to $9 billion from Brookfield.
A weekly roundup of AI agent developments including payment frameworks, cloud infrastructure, CRM agents, and security incidents.
A detailed guide on deploying n8n AI agents in production, covering Human-in-the-Loop, deployment checklist, and strategic context from SAP investment.
The article analyzes AMD's improved software and hardware prospects against Nvidia, highlighting key risks and potential market share gains.
TechCrunch provides news coverage on artificial intelligence and machine learning, including generative AI, large language models, and ethical issues.
Enterprise voice AI is transforming customer service by shifting repetitive tasks to machines, with industry-specific growth and regional dynamics, while trust and regulations shape adoption.
Enterprise Voice AI is reshaping customer service by shifting human labor to complex tasks, with growth led by finance and healthcare sectors, but trust and hybrid architectures remain key challenges.
AegisAI, Ropedia, and Paper each announce new funding rounds for AI-powered cybersecurity, Physical AI data infrastructure, and AI-native design platforms.
Nvidia dominates AI chip sales, AMD gains investor momentum with strong revenue and customer deals, while Intel's stock surge is tied to foundry turnaround, not AI chips.
South Korea's strong digital banking, government policy, and semiconductor infrastructure create favorable conditions for AI transformation in finance, impacting fraud detection, lending, and customer personalization.
A tool that compares token costs across 194 AI models from major providers, updated daily.
An article listing seven enterprise processes (invoice processing, customer support, employee onboarding, contract review, sales lead qualification, and more) that are ready for AI automation in 2026, with practical tips for implementation.
Article describes IA LAB, a monthly toolkit magazine for teachers to integrate AI into didactics, available from October 2026 to May 2027.
ScienceDaily aggregates AI research news covering energy-efficient chips, quantum computing, medical AI, and robotics.
A collection of news headlines about custom AI accelerators, chip partnerships, and AI infrastructure developments.
Guide on transitioning from developer to agent orchestrator using cloud VMs and agent swarms.
AI labs and models13 stories
South Korean President Lee Jae-myung met with Nvidia, OpenAI, Anthropic, and Broadcom CEOs in San Francisco to discuss AI investment and partnership, with Nvidia announcing a 500 billion dollar SK Group partnership and an AI Frontier Lab in Korea.
The AI Week in Review covers major model releases including Anthropic's Claude Opus 5, Poolside's Laguna S 2.1, and Google's Gemini 3.6 Flash, highlighting benchmarks and implications.
Tech giants urge US to avoid broad open-weight AI restrictions while Anthropic releases Claude Opus 5 with strong benchmark performance at competitive pricing.
Google's Gemini 3.5 Pro remains unreleased after 67 days past its initial promise, with rivals advancing during the delay.
OpenAI has launched GPT-5.6, a family of three models (Sol, Terra, Luna), and ChatGPT Work, a new agent that can manage apps, files, and long-running projects autonomously.
Punjab becomes the first Indian state to integrate AI as a core subject from grade 1 to 12 under the Sarbat.AI initiative, aiming to equip over 3.15 million students with future-ready skills.
A guide comparing all Claude models from Claude 3 to Opus 5 with specs, pricing, and recommendations.
AI companies are hiring philosophers to tackle alignment and ethical issues as technical hurdles are overcome.
A coalition of 25 companies urging no restrictions on open-weight AI models highlights a split, with OpenAI and Anthropic absent, reflecting a lobbying divide.
The article explains the shift from loop-based to graph-based AI agent engineering, covering methodology, parallel review patterns, and implementations with OpenClaw and Codex.
An analysis of top frontier AI models in July 2026, highlighting Anthropic's Claude 5 leadership and competitive landscape.
Grok Build can now create and run multi-agent workflows for complex tasks.
Perplexityの2026年におけるUI変更、モデル更新、新サービスCometとComputerの解説記事。
Policy and regulation7 stories
Microsoft, Meta, Nvidia, OpenAI, and others signed an open letter urging US policymakers to support open-weight AI models, countering potential restrictions on Chinese open-source models like Moonshot's Kimi K3.
Com o início das campanhas eleitorais, novas regras do TSE sobre uso de IA entram em vigor, proibindo deepfakes e exigindo transparência.
Delhi High Court denies interim injunction against OpenAI, ruling that training AI on copyrighted news content is fair dealing and that an injunction would harm AI development in India.
Several major AI companies including Nvidia, Microsoft, and Meta signed an open letter advocating for open-weight AI models, while OpenAI quietly added its support, as the White House accuses Chinese firm Moonshot AI of stealing intellectual property from Anthropic.
The Digital Omnibus amendment postpones some high-risk AI obligations but leaves many key rules in force from August 2026, so organizations should not halt compliance preparation.
Salamanca City Central School District halts its plan to introduce an AI-powered humanoid robot teacher after state officials raise concerns about student data privacy.
A federal appeals court rejected Trump's bid to halt a ruling blocking a $100,000 H-1B visa fee, upholding the lower court's decision that the fee was an unlawful tax.
Society, politics, and ethics5 stories
A coalition of religious leaders, labor unions, and local activists is emerging as a political threat to Trump due to AI concerns.
Vietnamese education officials discuss maintaining the current high school graduation exam model while improving organization and preparing for computer-based testing.
Former Education Secretary William Bennett and charter school CEO Jon Hage argue that AI in classrooms must be paired with ethical guardrails and character education to prevent harm to students.
Taylor Swift and other celebrities are fighting AI misuse of their likenesses through trademarks, petitions, and contract clauses after Meta's AI image tool sparked backlash.
A collection of opinion pieces and news briefs from The Guardian's AI section covering debates on datacenters, AI job impact, regulation, and ethical issues.
Consumer tech and culture2 stories
The article argues that TikTok's algorithm-driven, entertainment-first platform enables a uniquely effective form of affiliate marketing by reaching users before they express intent, with live shopping and AI-generated content further transforming the economics.
A detailed tutorial on using Midjourney in 2026, covering subscription plans, features of V8.1, and prompt writing techniques.
Labor and workplace2 stories
The article argues that AI-driven job displacement fears may be overstated, citing recent data and expert opinions questioning the pace of AI's labor market transformation.
An investigation reveals a concentrated impact of AI on young workers in entry-level roles, tempered by corporate AI washing and automation failures like Klarna's backfire.
Markets and macro2 stories
A summary of tech news including AI-fueled venture capital in South Africa and e-hailing industry changes.
An educational listing of defense stocks by category, with explanation of market drivers and risks.
Security and agent risk2 stories
Union Minister Piyush Goyal files police complaint after AI-generated deepfake video of his remarks is circulated to spread misinformation.
A blog post explaining AI scams, their rise, types, and how to protect against them, with a focus on practical advice.
About the RUDI Daily. Responsible Use of Digital Intelligence, daily. Compiled each day from same-day reporting across the web — every story links to its original publisher. How we build it →
Putting AI to work? RUDI helps organizations adopt AI responsibly — training, governance, and hands-on implementation. AI training · Consulting · Talk to us