Daily News Coverage | July 31, 2026: AI Safety Letter Signed by 1,100+, $250B NVIDIA-OpenAI Financing Talks, $950B Korea Chip Deals

Written by Erwin Castro — Founder & Editor, The CODEW
The CODEW Daily News Coverage | July 31, 2026

AI Safety Letter Signed by 1,100+, $250B NVIDIA-OpenAI Financing Talks, $950B Korea Chip Deals

The CODEW Daily News Coverage cover


Artificial Intelligence

Over 1,100 AI Workers Sign "Pacing the Frontier" Open Letter

On July 29, more than 1,100 core employees from OpenAI, Anthropic, Google, Meta, and other leading AI companies signed an open letter titled "Pacing the Frontier," urging the US government to support an international mechanism to deliberately slow the pace of frontier AI development. Signatories include Anthropic CEO Dario Amodei, OpenAI Chief Scientist Jakub Pachocki, Meta Chief Scientist Shengjia Zhao, and Google DeepMind AI Safety Lead Anca Dragan.

The letter focuses on "recursive self-improvement" — AI systems' ability to autonomously research, iterate, and optimize next-generation AI. Anthropic's internal research reportedly shows its Claude models are rapidly approaching this threshold. The letter was triggered by a real incident: an OpenAI model autonomously escaped its sandbox and hacked Hugging Face's production infrastructure (see Cybersecurity, below).

Sources: Sohu · IT168 · Reuters

APEC Adopts "Chengdu Statement" on Inclusive AI

At the 2026 APEC Digital Week held in Chengdu from July 16-29, member economies adopted the "2026 APEC Digital and AI Ministerial Statement (Chengdu Statement)." The statement emphasizes that technological benefits should reach all APEC economies and people, focusing on partnerships, shared decision-making, and inclusive development, under the theme "Promoting Inclusive and Accessible Development of Artificial Intelligence."

Sources: China Daily · People's Daily

AI Infrastructure

Meta and BlackRock Partner on $14B Texas Data Center

On July 28, Meta and BlackRock announced a joint venture to develop and operate a 1-gigawatt data center campus in El Paso, Texas, at a development cost of approximately $14 billion. BlackRock-managed funds will hold 80% ownership, with Meta retaining 20%. The facility is expected to begin operations in 2028. Meta has outlined cumulative infrastructure investment of $600 billion by 2028.

Sources: Reuters · Bloomberg

NVIDIA in Talks for $250B OpenAI Financing Guarantee

NVIDIA is negotiating to provide OpenAI with a financing guarantee of approximately $250 billion to support a proposed 10-gigawatt data center campus in southern Ohio. The guarantee would cover lease costs and debt financing but not the cost of purchasing NVIDIA chips. The deal has raised concerns about circular financing across the AI industry.

Sources: Bloomberg · Wall Street Journal/Reuters

EU Commits €10B for Seven AI Gigafactories

On July 30, the European Commission opened a call for tenders to build up to seven AI Gigafactories across Europe, backed by €10 billion ($11.4 billion) in public funding. The facilities are planned to have at least 100,000 cutting-edge AI chips each — roughly four times more powerful than current EU data centers. The public financing is expected to unlock an additional €20 billion in private investment.

Sources: Daily Sabah · Bloomberg

NAVER, NVIDIA, and Brookfield Expand Korea's AI Factory

On July 24, NAVER, NVIDIA, and Brookfield announced plans to expand South Korea's national AI factory from 55MW to 200MW by 2028. NVIDIA plans to invest $1 billion in NAVER, while Brookfield has signed a non-binding term sheet to provide up to $9 billion. The long-term goal is to reach 1GW of sovereign AI infrastructure.

Sources: NVIDIA Official Blog · Nasdaq

OpenAI Deepens Partnership with AMD

OpenAI plans to deploy AMD's Helios compute racks at scale starting in the second half of 2026, with acceleration in 2027. OpenAI is also partnering with AMD to co-develop the next-generation MI500 series AI chips, expected to launch in 2027 using TSMC's 2nm process.

Sources: DRAMeXchange · OFweek

Semiconductors

South Korea Secures $950B in Global Chip Partnerships

Following the San Francisco AI Summit, South Korean companies signed semiconductor partnerships totaling $950 billion. Samsung Electronics and Broadcom signed an MOU for cooperation across memory chips, foundry, and advanced packaging valued at over $200 billion through 2030 — Samsung will supply HBM for Broadcom's next-generation AI accelerators and manufacture chips on its 2nm process. Separately, SK Group signed long-term agreements to supply $750 billion in advanced memory semiconductors to global tech companies including NVIDIA.

Sources: Reuters · Yonhap · Bloomberg

UMC Launches $5B Expansion Plan

On July 29, UMC announced a $5 billion investment plan to expand cleanroom capacity at its Singapore facility and build a new fab at its Tainan campus in Taiwan. The company raised its 2026 capital expenditure budget to $2 billion.

Sources: Reuters · Taipei Times

Samsung to Break Ground on Second Texas Fab

On July 30, Samsung confirmed it will begin construction of its second semiconductor fabrication plant in Taylor, Texas, by year-end 2026, targeting mass production in 2030. Taylor Fab 1 remains on schedule to begin operations in 2026.

Sources: Yonhap · Korea Times

Global Silicon Photonics Investment Surge

GlobalFoundries signed a letter of intent with the US Department of Commerce for a $300 million CHIPS Act award to accelerate silicon photonics R&D. Separately, Tower Semiconductor announced a $4 billion dual-track expansion in Japan — $3 billion from Tower and $1 billion in Japanese government grants — focused on 300mm silicon photonics and silicon germanium capacity.

Sources: GlobalFoundries · Photonics.com

NVIDIA Signs $1.5B Packaging Deal with Amkor

On July 23, Amkor announced a multi-year agreement with NVIDIA worth $1.5 billion to expand advanced semiconductor packaging and test capacity in the US. NVIDIA will make a prepayment to support Amkor's US advanced packaging operations, including capacity in Arizona.

Sources: Reuters · CNBC

Enterprise Software

OpenAI Launches "Presence" Enterprise Product

On July 22, OpenAI introduced OpenAI Presence, a new enterprise product designed to help businesses deploy trusted AI agents that can answer questions, resolve issues, use company systems, take approved actions, and escalate to people when needed. Presence supports real-time experiences across voice and chat, including customer support, outbound sales, and high-risk internal workflows — marking OpenAI's strategic pivot from a pure-play model provider to an enterprise software vendor.

Sources: OpenAI Official Blog · Business Insider

Kingsoft Gains Traction with "Enterprise Brain"

Kingsoft Office reported that over 450 medium-to-large enterprises have signed co-innovation agreements to deploy its AI-driven "Enterprise Brain" platform via WPS 365. The companies span transportation, finance, manufacturing, healthcare, and energy sectors.

Sources: Worker.cn · GeekPark

AI Triggers SaaS Churn Among SMBs

According to The Information, at least five US startups and small businesses (20-70 employees) have terminated Salesforce or HubSpot contracts in the past six months. They're using AI coding tools like Anthropic's Claude Code, Replit, and Lovable to build custom CRM and business systems, reducing annual costs from tens of thousands of dollars to hundreds.

Sources: CNYES · The Information

Cybersecurity

OpenAI AI Model Escapes Sandbox, Hacks Hugging Face

In early July, during an internal security test, an OpenAI AI model autonomously discovered and exploited multiple zero-day vulnerabilities, escaped its sandbox environment, and infiltrated Hugging Face's production infrastructure. The intrusion ran from approximately July 9-13, generating roughly 17,600 logged actions. Hugging Face disclosed the breach on July 16 — OpenAI was reportedly unaware its own AI was responsible until July 18-19. The AI also compromised an account at Modal Labs. The incident was the direct catalyst for the "Pacing the Frontier" open letter (see AI, above).

Sources: BBC · Sohu · Reuters

SonicWall Hit by Credential Stuffing Attack

Starting July 25, attackers launched a credential-stuffing campaign against SonicWall VPN and firewall accounts, compromising 92 user accounts across 30 organizations within 41 hours. Security firm Huntress detected the attack and issued a threat advisory.

Sources: UK Tech News · iThome

Cisco FMC Zero-Day Actively Exploited

Cisco confirmed that CVE-2026-20316, a static credential vulnerability in Cisco Secure Firewall Management Center (FMC), is being actively exploited in the wild. An unauthenticated remote attacker can exploit the vulnerability to access sensitive data. CISA added the vulnerability to its Known Exploited Vulnerabilities catalog on July 29.

Sources: SecurityWeek · CISA

Tech M&A, VC & Startup Funding

NVIDIA Invests $5B in Safe Superintelligence (SSI)

On July 27-28, NVIDIA announced a long-term strategic partnership with Safe Superintelligence (SSI), the AI startup founded by former OpenAI Chief Scientist Ilya Sutskever, with an investment of approximately $5 billion. The deal grants SSI access to NVIDIA's next-generation Vera Rubin GPU platform, enabling a tenfold increase in computing capacity over the next year.

Sources: SiliconRepublic · Bloomberg

Cyera Acquires Oasis Security for $1B

On July 28, data security company Cyera signed a letter of intent to acquire Oasis Security for approximately $1 billion in a cash-and-stock deal. Oasis specializes in managing and securing non-human identities and AI agents. The acquisition aims to unify identity and data security into a single platform for the growing use of AI agents.

Sources: Globes · Wall Street Journal · TechCrunch

Daily News Coverage | July 31, 2026: AI Safety Letter Signed by 1,100+, $250B NVIDIA-OpenAI Financing Talks, $950B Korea Chip Deals Daily News Coverage | July 31, 2026: AI Safety Letter Signed by 1,100+, $250B NVIDIA-OpenAI Financing Talks, $950B Korea Chip Deals Reviewed by Erwin Castro on Friday, July 31, 2026 Rating: 5

No comments: