In Production

Weekly briefing

AI is already
in production.

The use cases your peers are not sending you — who deployed them, what changed in the P&L, and why it shipped in that market first.

This week’s copy

The use cases, in three lines.

Fashion's AI Era: Face Swaps and Digital Clones

AI-generated fashion models are revolutionizing the industry by offering brands cost-effective alternatives to traditional photo shoots. These digital models can reduce expenses by over 90%, posing a significant threat to human models and agencies. Brands like Diesel, Calvin Klein, and L'Oreal are exploring AI to modify images, create digital clones, and train AI tools using models' likenesses. However, this raises ethical concerns, as models risk losing control over their images and livelihoods. While some companies embrace AI, others proceed cautiously, and global regulations struggle to keep pace with these advancements.

AI Shoppers Revolutionize South Korea's Retail Industry

South Korean retailers are using AI-generated 'synthetic consumers' to test products and shopping experiences before launch. These virtual shoppers mimic real purchasing behavior using large language models and consumer data. Major companies like Lotte Members are developing personalized AI twins that analyze purchasing data from 45 million users. E-commerce platforms like Coupang and Naver are using the technology to simulate online shopping journeys. Intellicia, a synthetic-consumer provider, claims a 91.2% accuracy match with real consumer surveys in benchmark testing.

Google CEO Sundar Pichai Envisions Search Absorbed Into AI Agent Layer

Google CEO Sundar Pichai discussed plans to deeply integrate AI agents into Search, describing a future where an 'agentic layer' unifies user experiences. He explained that Google is developing shared agent infrastructure for both enterprise and consumer applications, with tasks increasingly performed by AI rather than just providing answers. Pichai noted this shift is user-driven, with agentic queries steadily increasing. The vision involves Gemini AI acting as a personal assistant across Google products, maintaining context 24/7. This represents a potential radical shift from current search paradigms, raising questions about the future of human referral traffic and websites.

Walmart Embraces AI Shopping Agents While Amazon Blocks Them

Walmart is launching an Agent Gateway to facilitate AI-powered shopping, contrasting Amazon's recent blocking of AI agents. The system allows approved AI agents to make purchases on customers' behalf using the Model Context Protocol. Walmart introduces Agent Profiles for transaction control, letting users set spending limits and approve purchases. The retailer maintains standard return policies even for agent-initiated purchases, addressing consumer concerns about AI errors. Walmart's approach differs from many e-commerce sites that have restricted AI agent access entirely.

AI Helped Find a Critical Security Flaw Affecting 70% of the Internet

Researchers from OpenAI, Wiz, and Calif used AI to uncover a critical security vulnerability called PHP2Shell, which affects 70% of the internet. The flaw allows attackers to remotely execute code, disrupt services, and expose sensitive data like passwords. AI significantly accelerated the discovery process, completing in a week what would have taken months manually. The PHP Foundation has released a patch, urging website operators to update their systems immediately. The vulnerability is compared to the 2014 Heartbleed bug, marking a potential turning point in cybersecurity.

Grain Uses AI to Streamline Operations and Save Time

Singapore-based F&B company Grain is leveraging AI to transform its operations, saving up to 20 hours a week. The company, founded by Yong Yi Sung, focuses on using technology to solve everyday challenges in the food and beverage industry. Grain’s stewarding team developed an AI system that predicts dirty dish volumes and optimizes manpower, reducing missed tasks and improving efficiency. Employees are encouraged to experiment with AI tools for tasks like summarizing information and automating repetitive work. Grain’s approach emphasizes continuous improvement and practical applications of AI, supported by government initiatives like Enterprise Singapore.

Show More Applications

Subscribe to the briefing

Open desk

Already running AI in the real world? Send it to the editors

Models

The models, in three lines.

GLM-Legal: Advancing Agentic Legal Intelligence

GLM-Legal is a large language model designed to handle complex legal tasks with precision and efficiency. It excels in parsing specialized contracts, merger filings, and judicial opinions, matching the performance of top closed-source models while operating on a leaner parameter budget. The model has been rigorously tested across real-world legal workflows, achieving a 19.2% all-pass rate on Harvey LAB and parity on LegalBench and PLawBench. GLM-Legal's training methodology focuses on domain-specific mid-training and post-training reinforcement learning, ensuring it can handle high-stakes legal environments. The model will soon be available on AutoClaw, a desktop-native agent platform, to assist with tasks like contract review, case preparation, and compliance research.

Meta Claims New AI Models Rival OpenAI and Anthropic

Meta has introduced two new AI models, Wonder and Wonder Full, which it claims are on par with OpenAI's ChatGPT and Anthropic's Claude. CEO Mark Zuckerberg stated that these models represent a significant advancement for the company. Meta has been investing heavily in AI, including a $14 billion investment in Scale AI and the formation of Superintelligence Labs. The new models will soon be available in Meta's personal agent app, Muse, for paid subscribers. Meta's AI spending could reach $145 billion by 2026 as it competes with leading AI labs.

Meta Introduces Muse Wonder Models for Advanced AI Tasks

Meta has launched Muse Wonder and Muse Wonder Full, its most advanced AI models to date. These models are designed to handle complex personal and professional tasks, such as navigating insurance claims, understanding leases, and building apps. Muse Wonder is available now for developers via the Meta Model API and Muse Code, while Muse Wonder Full is accessible to Muse Code subscribers. The models feature Active Memory and Active Focus, enabling them to retain relevant information across conversations and persist through challenging tasks. Meta emphasizes safety, ensuring the models avoid sharing private information and reject harmful requests.

Huawei to Launch New Kirin 7, 8, and 9 Series Chips Soon

Huawei is reportedly preparing to introduce new Kirin 7, 8, and 9 series smartphone chips later this year, according to a leak from Weibo tipster @DigitalChatStation. The new chipsets are expected to cover all categories from entry-level to flagship devices, marking a significant expansion of Huawei's chip lineup. The Kirin 7 series, in particular, could debut with Huawei's rumored Xingyao (Star Shine) sub-brand series. The company recently launched the Kirin 9050 Pro chip, which features major improvements over its 2025 predecessors. While the exact launch timeline remains unclear, some chips may appear in devices like the Huawei Enjoy 100 series by late 2026.

GPT-6 and Intelligent UI Now Available for Everyone

OpenAI has launched GPT-6, the latest iteration of its AI model, making it accessible to over 1.2 billion weekly ChatGPT users. GPT-6 introduces Intelligent UI, a feature that enhances responses with interactive and visual elements, making complex topics easier to understand and everyday tasks more efficient. The model can now generate tools like calculators or bill splitters directly within conversations. GPT-6 also improves response speed and accuracy, particularly for web searches, and includes enhanced safety measures to prevent misuse. The rollout begins today for ChatGPT Plus, Pro, Business, and Enterprise users, with Free and Go tiers following tomorrow.

Anthropic Launches Claude Haiku 5.5 AI Model with Major Speed and Cost Improvements

Anthropic has released Claude Haiku 5.5, its fastest and most cost-effective small AI model to date. The new model offers 75% lower costs compared to its predecessor while maintaining strong performance for high-volume tasks like summarization and classification. Haiku 5.5 features adjustable effort settings, allowing users to optimize for either cost or intelligence. Early adopters report significant latency reductions and improved accuracy in business applications. The release coincides with price reductions for Sonnet 5.5 cache reads and new API credits for Max and Team subscribers.

Show more models

Subscribe to the briefing

News

What the desks are reading.

Malaysia and China Universities to Establish Joint AI Research Lab

Universiti Malaya (UM) in Malaysia and Tsinghua University in China are collaborating to establish a joint artificial intelligence (AI) research laboratory at UM. The lab aims to advance AI technologies, develop research talent, and address global challenges. This partnership reflects a broader ambition to position Asian universities at the forefront of technological innovation and knowledge creation. Additionally, the collaboration includes an annual academic symposium and joint cultural initiatives.

Hong Kong Questions HSBC Over Decision to Establish AI Hub in Singapore

The Hong Kong Monetary Authority has questioned HSBC about its decision to establish a global AI center of excellence in Singapore rather than Hong Kong. The center, announced in July, aims to recruit over 100 AI specialists to enhance wealth management and payments technology. HSBC, under CEO Georges Elhedery, has been focusing on core markets, including Hong Kong, while also selling its Singaporean insurance business. Hong Kong authorities are keen to maintain the city's status as a financial hub, urging banks to locate more senior roles there. The move reflects ongoing competition between Hong Kong and Singapore as leading financial centers in Asia.

AMD to Acquire World Labs to Advance the Future of AI Compute

AMD has announced a definitive agreement to acquire World Labs, an AI research lab led by Dr. Fei-Fei Li, for approximately $8.2 billion. The acquisition aims to enhance AMD's capabilities in developing AI hardware, software, and systems by integrating World Labs' expertise in advanced AI models. This move is expected to strengthen AMD's position in shaping future AI infrastructure and supporting an open AI ecosystem. Following the transaction, Dr. Fei-Fei Li will join AMD as executive vice president and chief scientist. The deal is anticipated to close by the end of 2026, pending regulatory approvals.

Nvidia Releases AI Safety Software It Says Could Have Stopped Hugging Face Hack

Nvidia has launched a set of AI safety tools designed to prevent hacks like the one that targeted Hugging Face earlier this year. The company claims its software, including OpenShell and Sentry, could have contained the breach. These tools use hardware features on Nvidia's chips to detect and block rogue AI agents. Nvidia is collaborating with Arm Holdings and Intel to ensure compatibility across platforms. The release comes as OpenAI and Anthropic investigate similar AI agent breaches.

Singapore Launches National Semiconductor Brand Initiative, Inks R&D Deals

Singapore has launched a new national branding initiative called SG Semiconductor to promote its chip ecosystem globally. The initiative aims to attract high-value research and manufacturing investments by showcasing Singapore’s semiconductor capabilities. Key partnerships with global firms like Applied Materials, GlobalFoundries, and KLA were announced to deepen R&D collaborations. The initiative will focus on seven technology areas, including advanced packaging and silicon photonics. Singapore’s semiconductor sector currently produces about 10% of the world’s chips and 20% of global semiconductor manufacturing equipment.

Meta Launches AI Gadget Charm in Race for Post-Smartphone Hardware

Meta unveiled a handheld AI device called Meta Charm featuring a 5cm touchscreen and 5G connectivity. The gadget, designed to work with Meta's Muse AI assistant, will ship in December 2026 with pricing undisclosed. Meta also introduced camera-free Ray-Ban smart glasses to address privacy concerns, though audio recording capabilities remain. The company announced VR glasses priced at $1,299, set for spring 2027 release. These launches come as tech firms explore AI-powered devices that could potentially replace smartphones.

Show more news

Subscribe to the briefing