New Zealand Moves to Ban Social Media for Children Under 16

 New Zealand Prime Minister Christopher Luxon said his party will introduce legislation aimed at preventing children under 16 from using social media, with platforms potentially facing fines of up to 10% of their global revenue if they fail to comply.

Under the proposed bill, social media companies would be required to take reasonable measures to verify users' ages. These measures could include checking existing account information, using facial age-estimation technology and accepting digital identity documents.

Luxon said the government could not ignore the potential harm social media is causing young people, pointing to exposure to harmful content, addictive technology and online pressures. He said these issues can affect children's family relationships, mental wellbeing, sleep and education.

However, the proposed legislation faces uncertainty in Parliament. New Zealand First, one of the parties in Luxon's coalition government, has said it will not support the bill.

New Zealand First leader and Foreign Minister Winston Peters criticized the proposal, arguing that legislation restricting children's access to social media could lead the country down an undesirable path. He also described Australia's similar policy as a "colossal failure" and said parents should ultimately be responsible for deciding whether their children use social media.

Australia became the first country to introduce a nationwide ban on social media access for children under 16 in December. The rules cover major platforms including TikTok, YouTube, Instagram and Facebook.

Luxon acknowledged that enforcing the proposed restrictions would not be easy and that some children could find ways around the rules. However, he argued that the potential benefits make the effort worthwhile.

The proposed legislation will now face parliamentary scrutiny, with its chances of passing depending on whether the government can secure enough support from coalition partners and other lawmakers.

Xiaomi Unveils New Xring O3 Chip, Taps TSMC for 3nm Production

 Xiaomi has unveiled its latest in-house smartphone processor, the Xring O3, as the Chinese smartphone maker seeks greater control over key components and less reliance on external chip suppliers.

The new processor marks another step in Xiaomi’s effort to develop its own semiconductor technology, following the launch of its first proprietary chip, the Xring O1, in 2025. The move puts Xiaomi alongside major rivals including Apple, Samsung Electronics and Huawei, which have also invested heavily in developing their own smartphone processors.

TSMC to Manufacture Xring O3

Taiwan Semiconductor Manufacturing Co. (TSMC) will manufacture the Xring O3 using its advanced 3-nanometre production technology, according to two people familiar with the matter.

Developing processors in-house could give Xiaomi greater control over product features while strengthening its negotiating position with chip suppliers.

One source said the Xring O3 is expected to power Xiaomi’s upcoming flagship foldable smartphone, with the company targeting shipments of between 200,000 and 300,000 units.

Xiaomi’s growing presence in the foldable-phone market could increase competition with Huawei, which remains a dominant player in China’s premium foldable segment.

Huawei shipped about 1.6 million foldable phones in China during the second quarter, giving it a 68% market share, according to research firm Smart Analytics Global. Honor held 13.7%, while Oppo accounted for 8.5%.

Smartphone Makers Develop Their Own Chips

Xiaomi’s investment in proprietary processors reflects a wider industry trend as smartphone manufacturers look for ways to differentiate their devices and reduce dependence on suppliers such as Qualcomm and MediaTek.

The Beijing-based company restarted development of large, in-house chips in 2021 and has pledged to invest at least 50 billion yuan ($7 billion) over the next decade in the effort.

Xiaomi said last week that cumulative shipments of devices powered by the Xring O1—including smartphones, tablets and smartwatches—had exceeded 1 million units since its launch.

Sources said Xiaomi has sold around 150,000 smartphones equipped with the Xring O1 since its May 2025 launch.

Rising Memory Costs Add Pressure

The chip push comes as smartphone makers face weaker global demand and rising memory and component costs, which are contributing to higher device prices.

Xiaomi sold about 65 million smartphones in the first half of 2026 at an average price of 1,329 yuan ($197.74), compared with 84 million units at an average price of 1,141 yuan during the same period in 2025.

Global smartphone shipments are forecast to fall 14% in 2026, according to research firm International Data Corporation, adding further pressure on manufacturers to control costs and strengthen their supply chains.

Elon Musk Suffers Record $350 Billion Weekly Wealth Loss After SpaceX Shares Plunge

 Elon Musk has suffered the largest one-week decline in personal wealth ever recorded, losing approximately KSh 45.3 trillion ($350 billion) after SpaceX shares fell sharply from their post-IPO peak.

Musk’s net worth dropped from about KSh 187.8 trillion ($1.45 trillion) to just under KSh 142 trillion ($1.1 trillion), according to Forbes. The decline represents nearly 30% of his wealth from its peak on June 16, 2026. Despite the loss, Musk remains a trillionaire.

SpaceX IPO Sends Musk’s Wealth Soaring

SpaceX went public on June 12, 2026, at an IPO price of KSh 17,483 ($135) per share. During the following three trading sessions, the stock climbed as much as 67%, lifting SpaceX’s market value to approximately KSh 233.1 trillion ($1.8 trillion).

The surge pushed Musk’s net worth above KSh 129.5 trillion ($1 trillion), making him the first person to reach that milestone. His fortune had been estimated at around KSh 91.6 trillion ($707 billion) just two days before the IPO.

At its peak, SpaceX reached a valuation of roughly KSh 375.6 trillion ($2.9 trillion). However, the stock subsequently declined for three consecutive sessions, wiping out about KSh 120.2 trillion ($928 billion) in market value and reducing the company’s valuation to approximately KSh 259 trillion ($2 trillion).

By Monday, SpaceX shares had closed at $154, remaining above their IPO price but well below their post-listing high.

Why Did SpaceX Shares Fall?

The steepest one-day decline came after SpaceX announced plans for an investment-grade bond offering worth KSh 2.59 trillion ($20 billion). The company intends to use the funds to support its artificial intelligence expansion.

The announcement raised concerns among investors about SpaceX taking on additional debt, contributing to the sharp sell-off in its shares.

The dramatic reversal highlights the volatility that can accompany newly listed technology companies, particularly those that reach exceptionally high valuations soon after going public.

Musk Remains a Trillionaire Despite Record Loss

Musk’s rise to trillionaire status attracted worldwide attention and highlighted the extraordinary concentration of wealth among technology entrepreneurs.

Although his fortune has fallen sharply from its June peak, Musk remains above the KSh 129.5 trillion ($1 trillion) threshold. The SpaceX sell-off has nevertheless renewed debate over the sustainability of extremely high valuations and the risks investors face when newly public technology companies experience rapid price swings.

Novo Nordisk and AWS Expand Agentic AI Use Across Drug Discovery

 Novo Nordisk is expanding its use of artificial intelligence and Amazon Web Services (AWS) technology across drug discovery, with a focus on AI agents for identifying drug targets, designing therapies, and streamlining research workflows.

Under a newly announced agreement, AWS will become Novo Nordisk’s preferred cloud provider and strategic AI partner. The companies are also establishing a co-innovation hub at Novo Nordisk’s existing facility in London, bringing AWS engineers and pharmaceutical researchers together to develop AI-powered systems.

The initiative combines Novo Nordisk’s expertise in chronic diseases with AWS cloud infrastructure, AI services, and life sciences technologies. The companies aim to reduce delays between computational research and laboratory testing and ultimately shorten the time required to move potential treatments toward human trials.

Novo Nordisk and AWS Establish AI Research Hub

The London co-innovation hub will bring together Novo Nordisk research and development teams with AWS engineers, AI specialists, applied scientists, and professional services experts.

AWS said the arrangement is intended to reduce handoffs between computational analysis and laboratory research as potential drug candidates progress through early development.

Thilde Hummel Bøgebjerg, executive vice president of Enterprise IT & Quality at Novo Nordisk, said combining advanced AI capabilities with pharmaceutical expertise could help accelerate the development of new medicines.

The companies said the hub will focus on shortening the journey from identifying a potential drug target to delivering a first dose to a human participant.

AI Moves Deeper Into Novo Nordisk Research

Novo Nordisk already uses AWS technology for a range of internal AI applications.

According to an AWS case study, more than 25,000 Novo Nordisk employees have used a generative AI platform built with Amazon Bedrock. The platform supports more than 2,500 use cases, including information retrieval and document creation for non-regulated processes.

One major deployment reportedly contains approximately 140,000 documents and handles more than 26,000 prompts each month.

The pharmaceutical company has also applied generative AI to clinical-study documentation. AWS said a system using Anthropic’s Claude 3.5 through Amazon Bedrock reduced the time needed to produce certain clinical documents by more than 90%.

Tasks that previously required teams of 40 to 50 people and could take up to 15 weeks were reportedly reduced to minutes for a team of three. Medical professionals remain responsible for reviewing and validating the AI-generated material.

The latest partnership takes these AI capabilities further into scientific research and drug development.

Amazon Bio Discovery Targets Drug Development

A key technology involved in the partnership is Amazon Bio Discovery, an AI service designed to support computational biology and drug-development workflows.

AWS says the platform provides access to more than 40 biological AI models. Researchers can use AI agents to select and coordinate different models for specific research tasks.

Organisations can also combine models hosted by AWS with proprietary models to create multi-step computational workflows.

Novo Nordisk plans to use the technology for several stages of drug discovery, including identifying potential drug targets, designing therapies, and analysing biological data.

The platform can generate and rank potential drug candidates before selected compounds are sent for laboratory synthesis and testing.

Results from those physical experiments can then be returned to the computational workflow for additional analysis and model refinement.

This creates a feedback loop between computer-based drug design and laboratory experiments, potentially allowing researchers to test and improve candidate treatments more efficiently.

AI to Connect Clinical, Genomic and Imaging Data

Novo Nordisk also plans to use Amazon Bedrock to develop AI applications capable of working across clinical, genomic, and medical imaging datasets.

Connecting these different categories of information could allow insights from early-stage research to influence later stages of development, including clinical-trial planning.

The approach could give researchers a more integrated view of biological and clinical information, potentially helping them identify patterns that are difficult to detect when datasets are analysed separately.

Novo Nordisk Expands Use of AI Agents

Another major component of the agreement is Amazon Bedrock AgentCore, AWS infrastructure designed to help organisations deploy and operate AI agents.

AgentCore allows AI agents to interact with enterprise data, connect with existing systems, and perform multi-step tasks.

Novo Nordisk has not disclosed the specific AgentCore capabilities it will use in individual research projects. However, the company said it plans to introduce the technology across both research and operational workflows.

Dan Sheeran, vice president and general manager of Healthcare and Life Sciences at AWS, said the partnership is aimed at using agentic AI to remove bottlenecks throughout drug discovery rather than simply analysing those challenges.

The expanded relationship represents a shift from Novo Nordisk’s earlier AWS deployments, which were primarily focused on employee-facing generative AI applications, toward scientific workflows directly connected to drug development.

AWS Engineers to Work Directly With Novo Nordisk Teams

AWS Forward Deployed Engineers will work alongside Novo Nordisk teams on systems created through the partnership.

The programme places AWS technical specialists directly within customer projects, allowing teams to develop and deploy AI systems tailored to specific business and research requirements.

AWS CEO Matt Garman said the companies aim to use AI to accelerate drug discovery, analyse complex scientific datasets, and apply research findings across different therapeutic areas.

The partnership builds on a broader commercial relationship between Novo Nordisk and Amazon that includes services such as Amazon Pharmacy, Amazon Ads, and One Medical.

Novo Nordisk Pursues AI Beyond AWS

Novo Nordisk’s AI strategy extends beyond its partnership with AWS.

The company has also worked with OpenAI on applying artificial intelligence to areas including drug discovery, manufacturing, and commercial operations.

Novo Nordisk is additionally using Denmark’s Gefion supercomputer for workloads involving AI model development and other data-intensive scientific research.

The company’s expanding partnerships reflect the growing role of AI and high-performance computing in pharmaceutical research.

AWS has separately announced life-sciences AI collaborations with companies and organisations including Flagship Pioneering and Johnson & Johnson.

AI Could Reshape the Drug Discovery Pipeline

The partnership between Novo Nordisk and AWS reflects a broader pharmaceutical-industry push to use AI throughout the drug development process.

By combining AI agents, biological foundation models, cloud computing, proprietary research data, and laboratory experiments, Novo Nordisk aims to create more connected research workflows.

The ultimate objective is to accelerate the identification and testing of potential treatments for chronic diseases while reducing bottlenecks in early-stage drug discovery.

Novo Nordisk CEO Mike Doustdar said the AWS partnership is intended to speed up drug discovery and broaden the use of AI across the company’s work on chronic diseases.

Google’s AI Health Coach to Integrate Abbott’s Continuous Glucose Data

Abbott and Google are preparing to connect continuous glucose monitoring data with Google’s Gemini-powered Health Coach, expanding the range of personal health information available to the company’s AI-driven coaching service.

Under a multiyear agreement, data from Abbott’s Lingo continuous glucose monitor will be integrated into the Google Health app. Users will be able to view glucose trends alongside information such as physical activity, sleep, and other wellness measurements.

Google Health Coach, powered by Gemini, will use Lingo data together with other health information to offer personalised guidance on areas including nutrition, exercise, sleep, and recovery. The AI coaching service is available through a Google Health Premium subscription.

Google cautions that Health Coach is not designed for medical use and that its AI-generated responses may sometimes be incomplete or inaccurate. Users are advised to verify information and seek professional medical advice when appropriate.

Google Expands the Data Available to Health Coach

The Abbott partnership adds another category of health information to Google’s growing Health Coach ecosystem.

Google announced in May that users in the United States could connect medical records to the Google Health app. These records can include laboratory results, vital signs, and medication information, which users can then share with Health Coach for questions, analysis, or summaries.

The Google Health app can also collect information from compatible wearable devices and third-party services through technologies including Health Connect, Apple Health, and Google Health APIs.

Google and Abbott expect the Lingo integration to begin rolling out through the Google Health app later this year.

Continuous Glucose Monitoring Meets AI Coaching

Lingo is an over-the-counter continuous glucose monitoring system intended for adults aged 18 and above who do not use insulin.

The wearable continuously monitors glucose levels throughout the day, while its companion app helps users understand how factors such as food, physical activity, and everyday behaviour correspond with changes in glucose levels.

According to US Food and Drug Administration documentation, Lingo measures glucose in interstitial fluid rather than directly from the bloodstream. Its sensor is placed under the skin on the back of the upper arm and uses an electrochemical process to measure glucose before sending readings to the Lingo app via Bluetooth Low Energy.

The app can display current glucose readings, trend indicators, and glucose graphs. The sensor can be worn for up to 14 days, although the Lingo system does not provide glucose or system alerts.

Lingo and FreeStyle Libre Serve Different Purposes

Abbott developed Lingo using technology from its FreeStyle Libre platform. The company launched Lingo without a prescription in the US in 2024, and the product is also available in the UK.

While Lingo shares sensor technology with FreeStyle Libre 2, the products have different intended purposes.

FreeStyle Libre 2 is designed for diabetes management, whereas Lingo is intended to help adults who do not use insulin understand how glucose levels can relate to nutrition, exercise, and everyday activities.

FDA documentation also warns users not to make medical decisions based on Lingo readings without consulting a qualified healthcare professional.

Abbott and Google Plan Metabolic Health Study

The companies are also preparing a large real-world study focused on metabolic health.

The research will combine continuous glucose measurements with wearable-device information, laboratory results, and survey data. Researchers will examine relationships between glucose levels, physical activity, sleep, wellbeing, and broader metabolic health.

Abbott and Google say insights from the study could help improve Google Health Coach and contribute to future features for Lingo.

The partnership therefore goes beyond simply displaying glucose readings. It could provide Google with additional real-world data for understanding how continuous glucose measurements interact with other aspects of daily health and behaviour.

Google Brings More Personal Health Data to Gemini

The Google Health app is designed to bring information from multiple health sources together in one place. Users can connect activity, fitness, sleep, vital-sign, and medical-record information from supported services and devices.

Google says users control which information they save and can enable or disable optional features. The company also provides options to export or delete health information and says Google Health data is not used for Google Ads.

With Lingo integration, continuous glucose readings will become another potential source of information for Health Coach’s personalised responses.

Google has also been expanding Gemini’s role across healthcare-related services.

Zocdoc recently announced that US users can search for healthcare appointments and book providers directly through the Gemini app. The connected service provides real-time appointment availability from a network of more than 200,000 healthcare providers spanning more than 200 specialties.

The move makes Zocdoc Gemini’s first connected-app partner focused on healthcare appointments, highlighting Google’s broader effort to bring health-related services and information into its AI ecosystem.

What the Abbott-Google Partnership Could Mean

The integration of continuous glucose data with an AI health coach represents another step toward combining different types of personal health information in a single digital platform.

By bringing together glucose trends, sleep, activity, medical records, and other wellness information, Google aims to give Health Coach a broader picture of users’ health and daily routines.

However, because the system relies on AI-generated recommendations and sensitive personal health data, Google’s warnings about accuracy and the need for professional medical guidance remain important.

The Lingo integration is expected to begin rolling out later this year, potentially giving Google Health Coach access to one of the most detailed sources of continuous metabolic data available from consumer wearable technology.

Samsung Unveils AI Foundation Models to Unlock Continuous Health Insights from Wearables

 Samsung Research America’s Digital Health Team has introduced two artificial intelligence foundation models designed to learn from health data collected by wearable devices. The research focuses on biosignals recorded by smartwatches, including heart activity, sleep patterns, and physical movement.

Samsung presented its broader Connected Care vision during the Health Forum at Galaxy Unpacked in July 2026. The company envisions a healthcare ecosystem focused on prevention, personalised insights, and connected services, supported by advances in health technology and collaboration with healthcare partners.

Samsung’s researchers see health foundation models as an important part of this approach, with the potential to deliver continuous health insights while operating efficiently on consumer devices.

Sharanya Desai, Head of Digital Health Algorithms at Samsung Research America, said the research provides technical foundations for health models capable of delivering efficient, precise, and continuous insights.

The company said it plans to continue developing health foundation models that can work with different biosignals and health features while operating directly on devices with limited sensors and computing resources.

Samsung Develops Two Health AI Foundation Models

Health foundation models use self-supervised learning to identify meaningful patterns in large collections of largely unlabeled biosignal data. Once trained, a single model can potentially be adapted for applications including biosignal analysis, biomarker discovery, health prediction, and classification.

Samsung’s research introduces two models with different approaches to understanding wearable health data: xMAE and HiMAE.

xMAE, or Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning, is designed to learn relationships between different physiological signals over time.

HiMAE, short for Hierarchical Masked Autoencoder, focuses on identifying health patterns across multiple time scales within wearable time-series data.

Samsung says xMAE was accepted at the International Conference on Machine Learning, while HiMAE was accepted at the International Conference on Learning Representations.

Together, the models are designed to capture relationships between physiological signals and the way those signals change over time.

xMAE Connects Continuous PPG Monitoring With ECG Data

Electrocardiograms, commonly known as ECGs, directly measure the heart’s electrical activity. ECG data can provide information about heart rate, heart-rate variability, and abnormal rhythms, including signals associated with conditions such as atrial fibrillation.

However, wearable ECG measurements generally require users to actively perform a measurement.

Photoplethysmography, or PPG, works differently. PPG sensors detect changes in blood flow and can continuously monitor users through wearable devices such as smartwatches.

Although ECG and PPG measure different physiological properties, both are connected to cardiac activity. Samsung’s xMAE model is designed to learn the timing relationship between the two signals.

During training, the model reconstructs missing sections of ECG data using information from PPG measurements. This allows the system to learn relationships between continuously collected PPG data and ECG signals.

Samsung trained xMAE using approximately 9,400 hours of ECG and PPG recordings.

The approach could eventually support the analysis of cardiovascular-health indicators using continuously collected PPG data, reducing the need for users to perform separate ECG measurements for certain applications.

Samsung reports that xMAE outperformed unimodal biosignal models and existing multimodal learning approaches in 15 of 19 evaluation tasks.

The tests included cardiovascular disease prediction, detection of abnormal test results, and sleep-stage classification. Samsung also said the model’s learned representations showed potential across different sensors, body locations, and data-collection environments.

HiMAE Studies Health Patterns Across Multiple Time Scales

Wearable health data can reveal different information depending on the period being analysed. Short data segments can capture rapidly changing events such as individual heartbeats, while longer periods can reveal patterns associated with sleep, exercise, and physical activity.

HiMAE is designed to handle these differences by using multiple encoders to analyse short- and long-duration segments separately.

The approach allows the model to determine which time scale is most relevant to a particular health-related task. For example, heart-rate analysis may benefit from shorter sequences, while sleep-related predictions may require information collected over longer periods.

Like xMAE, HiMAE is trained by reconstructing masked sections of wearable biosignal data. This allows it to learn physiological patterns even when large amounts of manually labelled health data are unavailable.

Samsung says the same pretrained model can subsequently support classification, numerical prediction, and data generation.

Samsung Targets On-Device Health AI

One of the most significant aspects of Samsung’s research is its focus on efficient AI that can run directly on wearable hardware.

Samsung reports that HiMAE achieved strong performance while using a smaller model than existing alternatives. The company also says the model can generate results in less than one millisecond on a smartwatch-class central processing unit.

Running AI directly on wearable devices could reduce dependence on cloud servers and allow health-related analysis to take place locally.

This approach could be particularly useful for applications that require frequent or continuous analysis of physiological data. It may also reduce the need for constant connectivity between a smartwatch and remote computing infrastructure.

Subbu Venkatraman, Head of the Digital Health Research Lab at Samsung Research America, said the research demonstrates the potential of foundation models to capture both relationships between different biosignals and their underlying temporal patterns.

Samsung believes these capabilities could eventually support the identification of health markers, predictive health classifications, and personalised guidance using data collected by consumer wearable devices.

Samsung’s Broader Connected Care Strategy

The new AI models form part of Samsung’s broader effort to expand the role of wearable technology in health and wellness.

By combining continuously collected biosignals with AI models capable of operating efficiently on devices, Samsung is exploring ways to move health monitoring beyond occasional measurements toward more continuous and personalised insights.

The company’s research remains focused on developing the underlying AI technology, but the models demonstrate how wearable data could potentially be used for a wider range of health-related applications.

As wearable sensors become more capable and AI models become smaller and more efficient, on-device health intelligence could become an increasingly important part of consumer healthcare technology.

Nintendo Set to Enter Indonesia With Switch and Switch 2 Launch in December

 

JAKARTA, Aug. 12 (Reuters) — Japanese gaming giant Nintendo plans to officially launch its Switch and Switch 2 consoles in Indonesia this December, marking the company's first major entry into Southeast Asia's largest economy.

Nintendo confirmed the plan to Reuters in an email, saying both the original Nintendo Switch and the newer Nintendo Switch 2 will become officially available in Indonesia in December. The company did not provide further details about its planned operations in the country.

Indonesia's Ministry of Creative Economy said a subsidiary of major conglomerate Salim Group will serve as Nintendo's official distributor. The move is expected to make Nintendo products more accessible through authorized sales channels.

Indonesia's Vice Minister of Creative Economy Irene Umar said in an Instagram post that official distribution could also help tackle copyright infringement involving Nintendo products. She added that Indonesia is working to make Nintendo's developer kit available to local game creators, potentially supporting the country's domestic gaming industry.

Nintendo consoles and game cartridges are currently sold in Indonesia largely through unofficial channels, making prices and distribution less transparent, according to local game developer Adam Ardisasmita.

Axton Salim of Salim Group was quoted by state news agency Antara as saying that the group's subsidiary plans to work with Nintendo to establish an official repair center and customer-service operations.

Indonesia has the largest gaming market in Southeast Asia, with annual spending estimated at around 30 trillion rupiah ($1.68 billion), according to the Indonesian Game Association. However, local developers account for only about 2.5% of total spending.

Nintendo introduced the Switch 2, its latest hybrid home-and-portable gaming console, in June 2025. The company expects to sell 16.5 million Switch 2 units during the business year ending March 2027.

Inside the Executive Moves Behind Google’s Major AI Reshuffle

 

Alphabet has begun shifting some artificial intelligence teams from Google DeepMind into its broader corporate organization, according to two people familiar with the changes, as the company pushes to strengthen its position in the increasingly competitive AI race.

Google co-founder Sergey Brin has also urged key AI employees to focus intensely on Gemini, the company’s flagship AI model, two sources familiar with his remarks told Reuters. His involvement reflects the growing pressure on Google to accelerate Gemini’s development as competitors continue to make advances.

During an April town hall attended by hundreds of employees, Brin encouraged researchers and engineers at Google DeepMind to move faster, one person familiar with the meeting said. The comments came as rival AI company Anthropic was advancing its Claude model and intensifying competition in the sector.

Google’s Gemini briefly gained an edge over competing models last November, but subsequent releases from Anthropic and OpenAI pushed the company back into a catch-up position. By August, Google had postponed its next major Gemini release by about two months after internal testing showed that the model still lagged competitors in areas including software coding, according to two people familiar with the matter.

Google Reshapes DeepMind Leadership

On August 5, Google announced a major leadership restructuring at Google DeepMind. Demis Hassabis stepped away from day-to-day management and became chairman, while his deputy, Koray Kavukcuoglu, took on greater leadership responsibilities.

The company also lost the two original technical co-leads of Gemini, who left to establish a new startup.

The leadership changes have prompted speculation about Google's strategy in the AI model race. However, interviews with seven people familiar with Gemini’s development suggest that the restructuring is closely connected to Alphabet’s broader effort to improve the performance and commercial impact of its AI models.

The changes also represent a further reduction in DeepMind’s independence. Google has gradually integrated the London-based AI research lab into its broader operations since acquiring it in 2014, seeking to use its technology across more products and generate greater commercial returns.

Some AI Teams Move Into Corporate Google

During an August 6 all-hands meeting, employees were informed that some teams previously operating within Google DeepMind would be transferred into Google's corporate organization, according to two sources familiar with the meeting.

The internal meeting and Brin’s recent remarks have not previously been reported. Other media outlets had earlier reported on potential organizational changes and Brin’s increased focus on coding efforts.

Google declined to comment on questions regarding Brin’s involvement, changes to DeepMind’s structure and concerns among employees.

Hassabis’ Changing Role

Under Hassabis, Google DeepMind pursued a broad range of AI research while simultaneously developing Gemini models that became increasingly important to Alphabet’s commercial strategy.

However, a Reuters special report last year found that Hassabis had resisted some initiatives that could have generated additional revenue for Alphabet or strengthened Google’s position in the AI competition. Hassabis’ supporters at Google previously told Reuters that he remained committed to the company's business objectives and that his emphasis on long-term research did not come at the expense of commercial priorities.

Kavukcuoglu Gains Greater Influence

Google CEO Sundar Pichai expanded Kavukcuoglu’s responsibilities in 2025, when the DeepMind chief technology officer was also appointed Google’s chief AI architect.

His mandate included integrating Gemini more deeply into Google’s products and services. Four people familiar with the organization told Reuters that, in the months that followed, some DeepMind leaders saw their influence decline as Kavukcuoglu emerged as a central decision-maker on Gemini’s development.

A Google spokesperson said Kavukcuoglu will have the final say on major decisions involving DeepMind, underscoring his increasingly influential role as Alphabet reorganizes its AI operations and seeks to accelerate Gemini’s progress.

Disney Partners with TikTok to Bring Iconic Movie and TV Content to Short-Form Videos

 

Disney and TikTok Launch First-of-Its-Kind Content Collaboration

Walt Disney and TikTok have announced a landmark partnership that will allow creators on the popular short-video platform to use official content from Disney's vast entertainment library in their videos.

The agreement marks the first collaboration of its kind between TikTok and a major traditional media company, combining Disney's globally recognized franchises with one of the world's largest creator communities.

Disney Characters and Clips Coming to TikTok Creators

As part of the partnership, TikTok creators will gain access to selected content from hundreds of Disney movies and television series.

The available library will include material from some of Disney's biggest brands, including Pixar, Marvel, Star Wars and FX, enabling creators to produce new short-form content using officially licensed characters, scenes and creative assets.

The initiative is designed to encourage fan engagement while giving creators more opportunities to produce content inspired by Disney's storytelling universe.

TikTok Videos to Appear on Disney+

The collaboration will also introduce a new feature on Disney+.

A curated collection of TikTok videos will be made available through a dedicated "Verts" section within the streaming platform, focusing on vertical mobile-friendly content aimed at younger audiences.

The move represents the first time TikTok videos will be officially distributed on another streaming platform, reflecting the growing influence of short-form video in the entertainment industry.

U.S. Pilot Program to Launch First

The companies said the partnership will initially launch as a pilot program in the United States over the coming months.

Following the initial rollout, Disney and TikTok plan to expand the initiative into additional international markets, although no timeline has been announced.

Financial details of the agreement were not disclosed.

Streaming Platforms Embrace Vertical Video

Disney's latest move highlights a broader industry trend as streaming services increasingly adopt vertical video formats popularized by social media platforms.

Several major streaming providers have already introduced similar features. Peacock and Paramount+ now offer vertical programming, while Netflix has expanded its mobile experience with short-form "Clips" that help viewers discover new movies and television shows.

The shift reflects changing viewing habits, particularly among younger audiences who consume much of their entertainment through smartphones.

Partnership Could Boost Disney+ Engagement

Beyond giving creators new tools, the agreement is expected to help Disney strengthen engagement with its streaming platform.

TikTok has become an important discovery engine for entertainment content. According to company data, users shared an average of 6.5 million movie- and television-related posts every day during the past year.

Research also suggests that nearly half of viewers have watched a film or television series after first encountering related content on TikTok, demonstrating the platform's growing influence over viewing decisions.

Disney Sees Fans as Creative Partners

Disney Chief Marketing and Brand Officer Asad Ayaz said the partnership reflects the company's belief that passionate fans play a vital role in extending the reach of its stories.

He noted that the collaboration creates a stronger connection between Disney's storytelling and the creativity of millions of TikTok users, allowing audiences to interact with beloved franchises in new and engaging ways.

Outlook

The partnership signals a new stage in the relationship between traditional entertainment companies and social media platforms. As short-form video continues to reshape how audiences discover and engage with content, Disney and TikTok are betting that combining professional storytelling with creator-driven content will help attract younger viewers and expand the reach of some of the world's most recognizable entertainment brands.

Meta Apologizes to Indian Government After Briefly Restricting Prime Minister Modi's Facebook Post

 

Meta Issues Formal Apology Over Moderation Error

Meta Platforms has apologized to India's government after a Facebook post by Prime Minister Narendra Modi was briefly restricted due to what the company described as an operational mistake.

Joel Kaplan, Meta's Chief Global Affairs Officer, confirmed on Wednesday that he had personally apologized to India's Minister of Electronics and Information Technology on behalf of the company.

According to Kaplan, the restriction was unintentional and resulted from an internal error rather than a deliberate content moderation decision.

Government Summoned Meta Executives

The incident prompted India's Ministry of Electronics and Information Technology to summon Meta representatives last month for an explanation.

Following the meeting, Meta acknowledged that the restriction had been applied accidentally and assured officials that the issue had been corrected.

The temporary removal of content posted by the country's prime minister drew significant political attention and renewed scrutiny of how global social media platforms manage high-profile accounts.

Separate Legal Action Adds to Pressure

The controversy has been accompanied by additional legal challenges for the company.

Last week, police in Hyderabad reportedly registered a case against Meta India's head, Arun Srinivas, over Facebook videos that allegedly portrayed Prime Minister Modi in an offensive manner.

The investigation comes as authorities continue examining content published on the platform.

Political Protests Intensify Online Debate

The content-related disputes have emerged during a period of heightened political tension in India.

Prime Minister Modi has faced widespread criticism across social media as large youth-led demonstrations spread across the country following allegations of leaked national examination papers. The controversy eventually led to the resignation of India's education minister.

The protests generated a surge of political commentary, satire and criticism online, increasing pressure on social media companies to balance freedom of expression with compliance with local regulations.

Zuckerberg Also Reportedly Expressed Regret

Indian media reports said Meta Chief Executive Officer Mark Zuckerberg also apologized to government officials during a separate meeting concerning broader issues affecting the company's platforms.

The discussions reportedly covered concerns related to child sexual abuse material, deepfake content and other operational shortcomings. Meta did not immediately comment on those reports.

India Tightens Oversight of Online Content

The latest developments follow action taken by the Indian government in July, when authorities directed Meta to remove advertisements and content promoting child sexual abuse material from Instagram.

Meta reiterated that it maintains a zero-tolerance policy toward child sexual abuse content and said it continues investing in stronger detection technologies and enforcement systems to identify and remove harmful material more effectively.

Growing Regulatory Scrutiny

The recent incidents highlight the increasingly complex relationship between Meta and Indian authorities as the government strengthens oversight of digital platforms.

With content moderation, misinformation and online safety becoming key regulatory priorities, global technology companies continue to face mounting pressure to ensure their platforms comply with India's evolving digital rules while maintaining transparency in their moderation decisions.

AMD Shares Slide as AI Revenue Outlook Fails to Meet Lofty Investor Expectations

 

Investors Seek Bigger Returns From AI Boom

Advanced Micro Devices (AMD) shares fell sharply on Wednesday after the company's latest revenue forecast failed to satisfy investors who expected stronger evidence that the massive wave of artificial intelligence spending would drive even faster business growth.

The chipmaker's stock dropped 6.6% to $482.53 in early trading, putting nearly $59 billion of its market value at risk as investors reacted to guidance that, while above Wall Street estimates, fell short of the market's elevated expectations.

Nvidia Gains Another High-Profile Win

Investor sentiment toward AMD weakened further after Elon Musk confirmed that SpaceX would build its AI computing infrastructure exclusively with Nvidia processors. The announcement reinforced Nvidia's dominant position in the AI hardware market, helping its shares rise 3.3% during early trading while increasing competitive pressure on AMD.

The decision highlights the ongoing battle among leading semiconductor companies to secure large-scale AI customers as demand for advanced computing power continues to accelerate.

High Expectations Become AMD's Biggest Challenge

AMD has made significant progress in expanding its AI business, recently securing major agreements with AI startup Anthropic and digital infrastructure company Core Scientific. Those partnerships boosted confidence that the company could narrow the gap with industry leaders Nvidia and Intel.

However, analysts believe those achievements also raised investor expectations to exceptionally high levels.

Bernstein analyst Stacy Rasgon said market optimism had increased following Intel's recent earnings report, leaving investors with an already bullish outlook before AMD released its latest results.

Revenue Forecast Beats Estimates but Fails to Excite

AMD expects third-quarter revenue of approximately $13 billion, with a possible variation of $300 million. The guidance comfortably exceeded analysts' average forecast of $12.52 billion, according to LSEG data.

Despite beating consensus estimates, investors were looking for a stronger outlook given the rapid expansion of AI infrastructure spending across the technology industry.

Analysts Say the Company Faced an Extremely High Bar

Analysts at TD Cowen described AMD's latest earnings and outlook as fundamentally solid but noted that the company faced unusually high expectations following its recent AI customer wins and a strong rally in its share price.

The firm suggested that even positive financial results were unlikely to satisfy investors unless AMD demonstrated an even faster pace of growth.

Supply Chain Challenges Remain a Key Risk

Analysts at J.P. Morgan also warned that supply constraints could continue to limit AMD's expansion.

They pointed to heavy dependence on Taiwan Semiconductor Manufacturing Co.'s advanced N3 production process and CoWoS chip-packaging technology, both of which are expected to remain under significant demand through at least 2027.

Any prolonged bottlenecks could restrict AMD's ability to meet growing customer demand for AI accelerators and data-center processors.

Lisa Su Remains Confident in Long-Term Growth

Despite the market reaction, AMD Chief Executive Lisa Su expressed confidence in the company's long-term outlook.

She said the company expects data-center revenue to more than double by 2027 while forecasting overall revenue growth that exceeds its previously announced long-term target of more than 35%.

The data-center division continues to be AMD's fastest-growing business, with quarterly revenue more than doubling to $6.72 billion, surpassing analysts' expectations and reflecting strong demand for AI-focused chips.

Outlook

Although AMD continues to deliver solid financial growth and expand its presence in the AI market, the latest share decline shows that investors now expect exceptional performance rather than simply good results. With Nvidia maintaining its leadership and supply constraints still a concern, AMD faces increasing pressure to convert AI momentum into sustained earnings growth in the coming quarters.

Samsung and SK Hynix Explore Chinese Chipmaking Equipment to Reduce Future U.S. Trade Risks

 

South Korean Chipmakers Explore Backup Technology Options

Samsung Electronics and SK Hynix are reportedly evaluating semiconductor manufacturing equipment developed by Chinese supplier Advanced Micro-Fabrication Equipment (AMEC) as part of contingency planning for their operations in China.

According to sources familiar with the matter, the two memory chip giants have been testing AMEC's etching systems for nearly two years. The evaluations began as uncertainty grew over whether future U.S. export controls would continue allowing Western chipmaking equipment to be supplied and supported at foreign-owned semiconductor plants in China.

Both companies, however, have denied conducting tests of AMEC equipment for use in their Chinese manufacturing facilities.

Precaution Rather Than Immediate Adoption

People familiar with the discussions emphasized that the evaluations do not indicate plans for large-scale deployment of Chinese equipment. Instead, the testing is intended to provide an alternative option should future U.S. restrictions make it more difficult to service, repair or replace existing Western manufacturing tools.

The strategy reflects growing concern among global semiconductor companies over the long-term impact of geopolitical tensions on supply chains and factory operations.

U.S. Export Controls Continue to Shape Industry Decisions

The United States has steadily tightened restrictions on semiconductor technology exports to China over the past several years.

In 2023, Samsung and SK Hynix received "Validated End User" (VEU) status from the U.S. Commerce Department, allowing them to import certain controlled chipmaking equipment into their Chinese factories without obtaining individual licenses.

That authorization was later withdrawn in 2025 before being replaced with an annual license permitting equipment imports during 2026.

Despite the temporary approval, industry sources say both companies remain cautious about the possibility that future U.S. regulations could expand beyond new equipment and also restrict maintenance services or spare parts for machinery already installed at their Chinese plants.

Chinese Suppliers Viewed as a Contingency Plan

Rather than expanding manufacturing capacity in China, Samsung and SK Hynix are reportedly evaluating Chinese suppliers as a backup solution to help maintain existing production lines if access to Western technology becomes more limited.

Samsung operates a major NAND flash memory manufacturing facility in Xi'an, while SK Hynix runs NAND production in Dalian and a DRAM fabrication plant in Wuxi. These facilities currently rely heavily on etching systems supplied by leading American companies, including Applied Materials and Lam Research.

Maintaining uninterrupted production remains a priority should geopolitical developments disrupt existing supply arrangements.

Major Opportunity for China's Semiconductor Equipment Industry

For AMEC and other Chinese semiconductor equipment manufacturers, securing approval from global leaders such as Samsung or SK Hynix would represent a major commercial milestone.

Although Chinese suppliers continue to lag international competitors in advanced lithography and certain inspection technologies, they have made significant progress in several manufacturing processes, including etching, deposition, wafer cleaning and planarization.

Their equipment has also become increasingly attractive because of its competitive pricing compared with Western alternatives.

Geopolitical Tensions Drive New Industry Dynamics

The reported evaluations also illustrate an unintended consequence of ongoing U.S. technology restrictions.

While Washington's export controls are intended to slow China's semiconductor advancement, they have simultaneously encouraged multinational chipmakers operating inside China to consider local suppliers as part of their long-term risk management strategies.

Even without immediate deployment, the ongoing assessments could strengthen the credibility of Chinese semiconductor equipment makers and gradually expand their role in the global chip manufacturing ecosystem.

Outlook

No decision has been made regarding wider adoption of AMEC's equipment, and both Samsung and SK Hynix continue to deny testing the technology for use in China. However, the reported evaluations underscore how geopolitical uncertainty is influencing procurement strategies across the semiconductor industry, with manufacturers increasingly seeking diversified supply options to safeguard production against future disruptions.

Shein Eyes $30B–$40B Valuation for Hong Kong IPO as Listing Plans Gain Momentum

 LONDON/HONG KONG, Aug. 4 — Global fast-fashion retailer Shein is targeting a valuation of $30 billion to $40 billion for its planned initial public offering (IPO) in Hong Kong, with the share sale expected to launch as early as mid-August, according to three people familiar with the matter.

The proposed valuation and launch schedule remain under discussion and could change depending on investor feedback from a series of meetings that began last week. The sources requested anonymity because the listing plans are confidential.

The expected valuation represents a significant decline from Shein’s peak private market value. The company was valued at $98.2 billion during a funding round in 2022 before falling to $64 billion in both 2023 and 2024 as growth slowed and market conditions became more challenging.

People familiar with the process said the company is placing greater emphasis on ensuring a stable post-listing share performance rather than pursuing the highest possible valuation. Some prospective cornerstone investors are reportedly advocating for a valuation closer to $30 billion to $32 billion, reflecting a more cautious view of the retailer’s current outlook.

As part of its IPO preparations, Shein has held meetings with institutional investors in New York, Boston, and San Francisco to gauge demand ahead of the offering.

Headquartered in Singapore but founded in China in 2012, Shein operates in approximately 160 countries, offering low-cost fashion products ranging from dresses to denim. The company has yet to officially disclose the IPO size, pricing, or final listing timetable.

Financial documents submitted with its draft Hong Kong prospectus revealed that Shein recorded a $99 million quarterly loss, reversing previous profitability. The decline was driven by slower sales following the U.S. government's decision to remove duty-free treatment for certain low-value imports, along with a $328 million accounting-related charge tied to the valuation of convertible preferred shares.

Market analysts say investor caution is being driven by both weaker financial performance and increasing uncertainty surrounding Shein’s business model, which relies heavily on shipping low-value parcels directly to consumers.

Kenny Ng, a securities strategist at Everbright Securities International, said the reduced valuation primarily reflects concerns over declining earnings and the potential impact of evolving tariff policies on cross-border e-commerce. However, he added that there appears to be limited room for the company's valuation to fall much further.

Shein has also faced mounting pressure from rising trade costs, stricter regulatory oversight, and intensifying competition in the global online retail market. Slower revenue growth and shrinking profit margins have raised questions about whether the company can maintain the rapid expansion that made it one of the world's largest fast-fashion retailers.

After unsuccessful attempts to pursue listings in New York and London, Shein is now moving forward with its Hong Kong IPO after securing support from Chinese regulators, marking a key step in one of the year's most closely watched public offerings.

Meta, OpenAI, Google and Anthropic to Meet Trump Advisers as AI Security Concerns Escalate

 WASHINGTON, Aug. 4 — Representatives from Meta, Anthropic, Google, and OpenAI are set to meet with advisers to U.S. President Donald Trump on Tuesday to discuss voluntary safety testing for advanced artificial intelligence systems, according to four people familiar with the matter. The meeting comes as concerns grow over the cybersecurity risks posed by increasingly powerful AI agents.

The discussions follow recent disclosures by OpenAI and Anthropic that some of their AI systems were able to penetrate the digital defenses of other organizations during internal testing. Those findings have sparked fresh debate among U.S. lawmakers and security experts about whether next-generation AI models could eventually be used to launch or assist sophisticated cyberattacks.

According to sources, the White House meeting will focus on the federal government's plans to evaluate the hacking capabilities of the most advanced AI models developed by American companies. Officials aim to better understand potential security risks before these systems become widely available.

In June, the Trump administration announced that leading AI developers would be asked to voluntarily submit their frontier AI models for government safety assessments up to 30 days before public release. The testing framework is designed to identify potential vulnerabilities and evaluate how capable the systems are in areas such as cybersecurity.

On Tuesday, five Democratic senators urged President Trump to work with Congress on legislation that would make such testing a permanent requirement for the most advanced AI models rather than relying on voluntary cooperation from technology companies.

In a letter, the lawmakers argued that the United States must strike a balance between protecting national security and maintaining its competitive edge in artificial intelligence. They warned against creating an unpredictable regulatory environment that could disadvantage American AI developers while making competing Chinese technologies appear more accessible and attractive to global users.

The White House said on Monday that the details of the voluntary testing framework had been finalized but did not indicate whether the guidelines would be released publicly.

The administration has provided limited public information about its response to the recent AI security incidents, saying only that officials are monitoring developments involving OpenAI's reported cybersecurity testing. OpenAI Chief Executive Sam Altman also met with White House officials last week, underscoring the administration's growing focus on AI governance and national security.

The planned meeting highlights increasing efforts by U.S. policymakers and major technology companies to establish safeguards for advanced AI as the technology evolves rapidly and its potential impact on cybersecurity becomes a central concern.

Trump Administration Plans Ban on Chinese Data Center Components to Boost U.S. Infrastructure Security

 WASHINGTON, Aug. 4 — The Trump administration is preparing new restrictions that could block imports of certain Chinese-made data center components, according to four people familiar with the discussions. The proposal is part of a broader effort to strengthen the security of the infrastructure powering the rapid expansion of artificial intelligence in the United States.

Officials at the U.S. Federal Communications Commission (FCC) are drafting rules that would prohibit imports of new Chinese optical transceivers—critical networking devices that transmit high-speed data through fiber-optic cables inside data centers. Sources said regulators are aiming to release the measure before the end of the year, although the proposal could still be revised or abandoned.

The initiative is intended to reduce cybersecurity risks linked to Chinese technology, including concerns that foreign-made networking equipment could be exploited to collect sensitive information, install malicious software, or interfere with operations at facilities that host AI computing systems.

People familiar with the discussions cautioned that the proposal remains under review and no final decision has been made. Even so, it reflects the administration's continuing push to reduce U.S. dependence on Chinese technology in sectors considered strategically important.

Divyansh Kaushik, an AI policy specialist at Beacon Global Strategies, said optical transceivers represent a potential security concern because they are essential to the operation of modern data centers. He noted that securing the supply chain early is increasingly important as AI infrastructure expands across the country.

The report lifted shares of several American optical networking companies. Lumentum, Coherent, and Applied Optoelectronics all posted strong gains as investors anticipated they could benefit if Chinese competitors face new import restrictions.

China Warns of Countermeasures

Neither the White House nor the FCC commented on the reported proposal. China's embassy in Washington criticized the potential restrictions, urging the United States to avoid targeting Chinese companies and warning that Beijing would take necessary action if its economic interests were harmed.

A ban would likely affect Zhongji Innolight, one of the world's largest manufacturers of optical transceivers. The company was added to the Pentagon's list of firms alleged to have links to China's military earlier this year, a designation that often signals the possibility of further U.S. restrictions. Innolight did not immediately comment on the report.

Industry analysts say new import limits could also increase costs for American cloud service providers, including Amazon Web Services, which rely on optical networking equipment for large-scale data centers. Companies may need to shift purchases toward U.S.-based suppliers such as Coherent and Lumentum if Chinese products become unavailable.

The proposed restrictions also reflect lessons learned from the Huawei case, where Chinese telecommunications equipment became deeply integrated into U.S. networks before Washington imposed sweeping sanctions. Removing that infrastructure has proven costly and time-consuming, prompting policymakers to take earlier action in emerging technology sectors.

According to market research firm Counterpoint Research, Innolight controls roughly 27% of the global data center transceiver market, making it one of the industry's largest suppliers. Although U.S. manufacturers produce competing products, analysts say they currently lack the production scale needed to fully replace Chinese suppliers. Research from the Foundation for American Innovation also notes that approximately 90% of Innolight's revenue is generated outside China, highlighting its significant presence in international markets.

Prysmian Expands U.S. Business with $3.8 Billion Atkore Acquisition

 

MILAN, Aug. 3 — Italian cable manufacturing giant Prysmian has agreed to acquire U.S.-based electrical products company Atkore in a deal valued at $3.8 billion, strengthening its presence in the United States as demand for electrification projects and AI-powered infrastructure continues to grow.

The acquisition marks Prysmian’s latest major investment in North America, following its $4.79 billion purchase of Texas-based Encore Wire in 2024 and the $1.15 billion acquisition of California’s Channell Commercial in 2025. More recently, the company secured a long-term agreement worth up to €5.5 billion with electronics manufacturer Molex to supply fiber-optic cables for data center projects.

North America now generates roughly 40% of Prysmian’s total revenue, making the region a key growth market for the Milan-headquartered company.

Chief Executive Officer Massimo Battaini said Atkore is a strong strategic fit that will significantly expand Prysmian’s scale and product offering in the U.S. market. He also indicated the company is already exploring additional merger and acquisition opportunities in other regions after successfully securing Atkore through a competitive bidding process.

Under the agreement, Prysmian will pay $95 per share in cash for Atkore, representing a premium of around 30% over the company’s previous closing share price.

Financing Plan

Chief Financial Officer Pier Francesco Facchini said the acquisition will be financed through a combination of funding sources. Around 20% will come from equity, including the possible sale of treasury shares, while more than 20% will be raised through hybrid financial instruments. The remaining 60% will be financed through debt.

Despite the announcement, Prysmian’s shares reversed early gains. After opening about 3% higher, the stock later traded roughly 3% lower during the session.

Stronger Product Portfolio

Atkore employs approximately 5,400 people worldwide and reported $2.85 billion in revenue for 2025, along with $386 million in EBITDA. The company manufactures electrical conduit systems, cable management solutions, and other products used in data centers, renewable energy projects, utilities, and transportation infrastructure, including railway networks.

Battaini said the two companies offer complementary products to many of the same customers, creating opportunities to provide bundled solutions and strengthen customer relationships.

Prysmian expects the transaction to deliver approximately $150 million in annual pre-tax synergies within three years of completion. The acquisition is expected to close by the end of December, subject to customary approvals.

On a combined pro-forma basis, the merged company would have generated approximately €22 billion in 2025 revenue and €2.7 billion in adjusted core earnings, further reinforcing Prysmian’s position as a global leader in electrical and connectivity solutions.

China Enhances Chip Design Protection with Tougher Intellectual Property Rules

BEIJING, Aug. 3 – China has introduced updated regulations to strengthen the protection of integrated circuit layout designs, reinforcing intellectual property rights for domestic semiconductor technology. The revised framework introduces stricter registration requirements and allows courts to impose punitive damages in cases of serious infringement, highlighting Beijing’s growing focus on protecting homegrown chip innovation.

The new regulations, approved by Premier Li Qiang on July 23, are scheduled to take effect on October 15, according to a joint announcement from the Justice Ministry and the China National Intellectual Property Administration published by state media on Monday.

Integrated circuit layout designs define the physical arrangement of components within a semiconductor chip. These layouts often represent years of engineering expertise and play a critical role in chip development, even for companies that focus solely on design rather than manufacturing.

Although the updated rules do not introduce export restrictions, they underscore China's increasing emphasis on securing strategically valuable technologies developed by domestic firms. The changes come as the country continues to strengthen its semiconductor industry amid ongoing technology-related restrictions imposed by the United States.

Recent reports have suggested that Chinese policymakers are also exploring broader measures to prevent key domestic technologies from being transferred abroad or acquired by foreign companies. Such proposals could reportedly include tighter oversight of advanced chips produced overseas using Chinese-developed designs.

Under the revised regulations, applicants seeking protection for chip layouts must demonstrate that their designs are the result of original creative work. They will also be required to submit declarations confirming originality and clearly identify which elements of their submissions are unique.

Authorities will have greater authority to reject applications that fail to meet these standards, while the updated rules provide stronger legal grounds for challenging registrations that should not have been approved.

According to Alfredo Montufar-Helu, Managing Director at Ankura China Advisors in Beijing, the stricter originality requirements and tougher registration standards are intended to reduce low-quality or bad-faith filings while ensuring that companies with genuine technological capabilities receive stronger protection.

The revised framework also expands legal remedies for intellectual property violations. Compensation may now be based on either the financial losses suffered by the rights holder or the profits earned by the infringing party, with courts empowered to award punitive damages in particularly serious cases.

In addition, the regulations clarify the legal procedures governing the licensing, transfer, and use of chip layout design rights as collateral. Organizations responsible for developing these technologies will also be required to provide fair rewards and compensation to individuals who contribute to their creation.