Brief

China's National Data Administration convened a meeting on June 4th to discuss improving data rules and fostering AI innovation. Participants emphasized balancing data development with security and advancing market-based reforms for data elements. The goal is to enhance the data supply system for AI, unlock data value, and drive AI development through data. AI

A significant portion of users and organizations are already deploying AI agents that operate autonomously without human oversight, according to recent reports. Despite this rapid adoption, many companies lack adequate governance frameworks to manage these agents, creating accountability gaps. This shift from AI as a content generator to an autonomous actor presents new risks, as decisions are executed before human review, necessitating a reevaluation of organizational responsibility. AI

Anthropic CEO Dario Amodei has proposed new policy frameworks for AI regulation, likening current governmental speed to a slow-moving tree. He argues that the rapid advancement of AI, exemplified by Anthropic's latest model, necessitates faster regulatory action. Amodei's proposals include empowering regulators to ground frontier models, establishing independent screening systems, and addressing potential job displacement through new economic models. AI

Tata Consultancy Services (TCS) has partnered with Anthropic to accelerate enterprise AI adoption. TCS will deploy Anthropic's Claude models to 50,000 employees across various departments, including engineering, finance, and marketing. This collaboration aims to scale AI implementation for clients. AI

Chinese firm Biwin has secured a significant memory chip supply deal worth US$1.86 billion, extending from mid-2026 to mid-2028. This multi-year agreement with an undisclosed supplier aims to guarantee capacity and delivery schedules, mitigating risks from market volatility. The contract's value surpasses Biwin's projected 2025 revenue, highlighting the company's proactive strategy to lock in upstream supply during the current memory upcycle. AI

Chinese satellite internet provider Spacesail is emerging as a competitor to SpaceX's Starlink, launching satellites and securing partnerships in markets where Starlink has faced regulatory or service issues. Spacesail aims to build a constellation of 15,000 satellites by 2030, having already launched 200 and secured deals with companies like Airbus and state telecom operators in various countries. This competition is particularly relevant as SpaceX prepares for a major public offering, with Spacesail's strategy focusing on countries where Starlink's market dominance has led to price increases or regulatory friction. AI

Tata Consultancy Services (TCS) has partnered with Anthropic to deploy Claude models to 50,000 employees across various departments. This collaboration aims to accelerate the adoption of enterprise-level AI for TCS's clients. Separately, JD MALL has introduced AI-powered robots for roles like greeters and stockers, with one model achieving over 90% success in stocking items. AI

The UK Treasury is reportedly undecided on its participation in a £1.7 billion ERP program it previously agreed to fund. This program is intended to boost the UK's AI and computing infrastructure. The delay in commitment raises questions about the government's support for critical technology initiatives. AI

A new methodology has been proposed to accurately account for the greenhouse gas emissions generated by AI inference services within corporate sustainability reports. This four-tier framework aims to provide a more precise estimation than current practices, which often omit these emissions or use overly broad economic factors. The proposed method utilizes GPU energy benchmarks and regional grid carbon intensities for direct estimation, with a fallback to spend-based economic factors for services lacking usage data. AI

Switzerland is set to vote on a referendum that could implement the world's first national population cap, potentially limiting the population to 10 million by 2050. Major companies like Roche and Nestlé, along with the Swiss business federation economiesuisse, fear this initiative will severely restrict access to international talent, thereby hindering economic growth and innovation. The proposal, supported by the Swiss People's Party, aims to limit immigration and population growth, but critics argue it would damage Switzerland's competitiveness and its ability to attract highly qualified foreign workers essential for research-intensive industries. AI

Anthropic CEO Dario Amodei has proposed a federal regulatory framework for AI, drawing parallels to the FAA's oversight of aviation, advocating for mandatory third-party testing and deployment holds on frontier models. Google DeepMind has released DiffusionGemma, an open-weight model that uses diffusion techniques for faster, on-device text generation, potentially revolutionizing local AI applications. Meanwhile, MassMutual's CIO shared the company's strategy of avoiding vendor lock-in with short contracts and a focus on model quality, which has reportedly yielded significant productivity gains. AI

Oracle's CFO announced that the company anticipates $70 billion in net capital expenditures for fiscal year 2027, with returns secured by long-term customer contracts. Separately, Hong Kong Stock Exchange data shows 90 biotech companies listed under Chapter 18A since 2018, raising over 142.9 billion Hong Kong dollars. The exchange's Chapter 18C, for specialized tech firms, has seen 10 companies list in the first five months of 2026, primarily in AI, robotics, and autonomous driving. AI

iFlytek has established a new subsidiary, Hefei Ciyuan Xinghuo Technology Co., Ltd., with a registered capital of 1 billion RMB. This move indicates a significant investment by iFlytek in AI development and applications, as the subsidiary's business scope includes AI software development and general AI systems. Meanwhile, former US President Trump has expressed his desire for top AI companies to share profits with the public, suggesting potential government stakes in these companies. AI

SpaceX has reportedly begun trading on the Nasdaq, with its opening price at $174, a 29% increase from its IPO price. In other news, Shell has temporarily suspended its stock buyback program due to securities regulations related to ARC Resources, with the plan to resume after a shareholder meeting on July 14th. Additionally, a 90s-born tech expert, Chen Yuson, has taken over as CEO of DingTalk, and Bill Gates is scheduled to testify before Congress regarding the Epstein case. AI

Anthropic is launching a new initiative called Claude Corps, dedicating $150 million to embed 1,000 AI fellows within nonprofit organizations across the United States. These fellows, earning $85,000 annually plus benefits, will focus on assisting nonprofits in leveraging AI, specifically Anthropic's Claude models, more effectively. The program aims to support early-career professionals and requires no specific degree, emphasizing practical application of AI skills. AI

Commercial data centers can tap into federal demand by meeting stringent US government requirements for uptime, security, and compliance. Operators must prove their facilities are ready, covering aspects like physical infrastructure, power, cooling, and network diversity. Additionally, they need to support agency-specific authorization processes, such as Authority to Operate (ATO) under the NIST Risk Management Framework or FedRAMP for cloud services. AI

A dispute between the Nashville Zoo and a proposed AI data center, DC BLOX, has intensified with the zoo filing a zoning appeal to overturn the data center's approved permits. The opposition has garnered over 330,000 signatures and celebrity support, leading to proposed legislation that could ban large data centers and impose strict siting rules. This local conflict mirrors a broader national trend of public opposition to data center development. AI

Shell has temporarily suspended its $3 billion stock buyback program due to restrictions under securities laws, which were triggered by the release of ARC Resources' shareholder circular. The buyback, initiated on May 7, 2026, will pause from June 12, 2026, until after the ARC shareholder meeting on July 14, 2026. Any unexecuted portion of the buyback is expected to be rolled into subsequent repurchase plans for 2026, pending board approval. In unrelated news, Tencent's AI efforts are shifting, with 'Yuanbao' reportedly losing favor to 'WorkBuddy'. AI

China has imposed a travel ban on the Philippine defense chief and his family, which analysts interpret as a strategic warning to Manila. This move, described as "killing the chicken to warn the monkeys," aims to deter other critics. While Beijing claims the ban is a direct response to remarks made by the defense chief, influential figures in the Philippines are calling for reciprocal sanctions to demonstrate that the nation will not be intimidated. AI

U.S. banking regulators are increasing their oversight of artificial intelligence use in the financial sector. The focus is on how financial institutions manage and deploy AI technologies, particularly concerning risk management practices. AI

The United States and Iran are reportedly nearing an agreement to reopen the Strait of Hormuz. This potential deal, which could be signed during the upcoming G7 summit, would extend their current truce by approximately two months. As part of the agreement, Iran would reopen the strait, and the US would lift its blockade on Iranian ports, paving the way for further negotiations on Iran's nuclear program. AI

Australia's Labor government plans to regulate the burgeoning data center and AI industries to avoid repeating past resource boom mistakes. Assistant Minister for the Digital Economy, Andrew Charlton, announced a proposed "triple lock" policy requiring new data centers to offset their energy demand with renewable sources, cover infrastructure costs, and support grid stability. While industry groups express concerns about the feasibility of immediate 100% renewable offsets, government officials and clean energy advocates argue that project timelines are improving and that legislative backing is crucial for effective implementation. AI

The FBI has dismantled 13 websites operated by Chinese intelligence services that were allegedly used to target Americans with security clearances. These sites, disguised as consulting firms, aimed to recruit or coerce individuals into divulging sensitive information, with some employing AI-generated content. China's embassy in Washington has denied the allegations, calling them fabricated and slanderous. AI

The U.S. House of Representatives has rejected a short-term extension for Section 702 of the Foreign Intelligence Surveillance Act, putting the critical surveillance authority at risk of expiring. This legislative deadlock stems from a dispute over President Trump's appointment of Bill Pulte, who lacks national security experience, as acting director of national intelligence. Democrats have stated they will not support a FISA extension unless Trump rescinds this appointment, while some Republicans believe the FISA authority is too important to let lapse over this personnel issue. AI

A lawsuit has been filed against Westchester County, New York, alleging its extensive license plate reader network constitutes an unconstitutional surveillance system. Civil rights groups claim the county amassed 1.6 billion plate scans without proper authorization and shared this data, including travel patterns, with over 50 agencies like ICE. The suit seeks to halt the program, citing privacy violations for millions of motorists. AI

A new academic paper proposes the creation of an "AI Legal Specialist" role to navigate the complex and rapidly evolving landscape of AI regulation globally. The paper argues that existing legal roles are insufficient to address the unique challenges posed by AI governance, citing comprehensive regulations like the EU AI Act and initiatives in various other countries. This distinct professional profile, defined by its juridical autonomy, would focus on the intersection of legal interpretation and AI governance, with a proposed competence architecture and performance indicators for standardization. AI

A new report from arXiv details significant challenges in applying UK Defence's AI assurance directive, JSP 936, in real-world scenarios. The analysis highlights eight key areas of difficulty, including evidence management, human-AI interaction, and ethical considerations. The paper concludes that while JSP 936 offers a foundational governance framework, its effective implementation requires further development in methods, guidance, and organizational capabilities to ensure safe and responsible AI adoption. AI

The Shanghai Stock Exchange has intensified its scrutiny of listed companies, conducting special investigations into 22 significant matters this week. This includes close monitoring of stocks with severe abnormal fluctuations, delisting process stocks, and high-premium funds like the Global Chip LOF and Caitong Fuxin LOF. Additionally, one case suspected of illegal and non-compliant activities has been reported to the China Securities Regulatory Commission. Separately, a company named Fuxin Technology reported a significant increase in its communication sector revenue, particularly for MicroTEC products used in high-speed optical modules, with its capacity utilization rate exceeding 95%. AI

Canada is enacting legislation to ban social media access for children under 16, unless platforms can demonstrate robust safety measures. This move aligns with a global trend of countries implementing stricter regulations to protect minors online. The proposed law, which includes creating a new Digital Safety Commission, aims to combat harmful content such as self-harm promotion, violence, hatred, and non-consensual intimate imagery. AI

In separate incidents, artificial intelligence has caused significant disruptions in legal and search contexts. In one case, lawyers in a court proceeding used AI to cite non-existent case law, leading to the judge's anger and the suspension of the trial. In another, a German court ruled that Google bears responsibility for any inaccuracies in its AI-generated search summaries, a landmark decision for AI accountability. AI

South Korean prosecutors have raided Samsung Electronics' headquarters as part of an investigation into alleged insider trading by personnel at its subsidiary, Rainbow Robotics. The individuals are suspected of illegally profiting approximately 3 to 4 billion Korean won by using non-public information during Samsung's acquisition of the company between 2022 and 2024. Financial regulators have transferred 16 individuals, including the president and CFO of Rainbow Robotics, to the prosecution. AI

The China Securities Regulatory Commission (CSRC) has approved the registration application for the initial public offering (IPO) of Chengdu Electronic Science and Technology Co., Ltd. (CETC Synergetic Technology) on the ChiNext board. Separately, ST Lihang announced that its stock will be suspended for one day on June 15, 2026, and will resume trading on June 16, 2026, with its delisting risk warning removed. Following this, the company's A-share name will change to Lihang Technology, and its daily trading price limit will increase from 5% to 10%. AI

Hong Kong has introduced a tax reform bill aimed at strengthening its position as a global wealth management hub. The proposed legislation, set to be reviewed by lawmakers on June 24, would offer salary tax exemptions on performance-linked bonuses for fund managers, private equity, and venture capital professionals. This initiative seeks to attract international talent and investment by making Hong Kong the first major Asian financial center to provide such tax relief on performance income, while also expanding tax exemptions to a wider range of investment products. AI

BOE Technology Group has been mistakenly included on a US Department of Defense list of Chinese military companies. The company asserts it is not a military enterprise and the designation is unjustified. BOE stated that its operations remain unaffected as the US restrictions only pertain to defense procurement and do not impede other business dealings. AI

China is rapidly expanding its nuclear power capacity with large-scale reactors, significantly increasing its fleet since 2016 and aiming to surpass the US and EU in installed capacity by 2030. This contrasts with the US and France, which have added few new reactors recently and are focusing on smaller, modular designs. Companies like Antares are developing microreactors, attracting investment from both government programs and tech giants hoping to power data centers, though these smaller reactors are still in early development stages. AI

Canada is proposing a new digital safety bill that would prohibit social media use for individuals under 16, with potential exemptions for platforms meeting specific safety standards. The legislation also aims to regulate AI chatbots by requiring companies to mitigate the risk of harmful content dissemination. This move follows Australia's implementation of a similar age restriction for social media users under 16. AI

Autonomous drones have reportedly been deployed in Ukraine, leading to confirmed casualties. Experts warn that these systems lack the ability to discriminate between combatants and civilians, posing a significant risk to non-combatants. The deployment raises ethical concerns about the use of AI in warfare. AI

Chinese authorities have identified 30 applications, including educational and e-commerce platforms, for violating personal information collection and usage regulations. These violations range from failing to disclose data collection rules and requesting unnecessary permissions to not accurately listing SDK data usage and lacking effective account cancellation features. Operators of the affected apps have been given 15 working days to rectify these issues, with further action to be taken based on their compliance. AI

Members of Congress are responding to growing public opposition against data centers, driven by concerns over environmental impact and energy consumption. Several legislative proposals have been introduced to restrict data center construction and protect communities from increased utility costs. One bill specifically aims to prevent companies from suing municipalities that reject data center applications and requires community benefit agreements for federal tax incentives. AI

Section 702 of the Foreign Intelligence Surveillance Act, a critical tool for U.S. national security, is at risk of lapsing due to political deadlock. Democrats are blocking its extension unless former President Trump withdraws his nomination of Bill Pulte as acting director of national intelligence, citing Pulte's lack of experience and temperament. House Speaker Mike Johnson and other Republicans are urging for the extension, emphasizing its role in thwarting terror plots and informing the president's daily briefing. A short-term extension is planned, but its passage is uncertain. AI

South Korea's Personal Information Protection Commission has fined e-commerce giant Coupang a record 624.7 billion won (approximately US$409 million) for a significant data breach. The commission found that inadequate security measures led to the exposure of personal data belonging to around 37.5 million users. This penalty is the largest ever imposed by South Korea for a personal data breach, highlighting the regulator's focus on data protection following rapid e-commerce growth. AI

Canada's privacy commissioner has found that Elon Musk's Grok AI violated federal privacy laws by failing to implement adequate safeguards. The investigation revealed that the AI tool generated over 6,000 sexual deepfake images per hour at its peak, with millions of such images created. These deepfakes primarily targeted women and children, leading to a formal finding of violation against X Corp and xAI. AI

A Florida man was wrongfully arrested due to an inaccurate facial recognition match from a system called FACES, operated by the Pinellas County Sheriff's Office. Despite the system indicating a 93% match, the man lived hundreds of miles away and had never visited the city where the crime occurred. The ACLU has filed a lawsuit against the involved agencies, alleging failures in police investigation and seeking damages and policy overhauls for face recognition use. AI

China Mobile Hong Kong will stop allowing its prepaid SIM cards to be activated in mainland China starting June 17, 2026. This move is seen by some as an attempt to block mainland users from legally accessing AI services through VPNs. The change affects all of China Mobile Hong Kong's prepaid SIM card products. AI

The Hong Kong Stock Exchange has seen significant activity under its Chapter 18A listing rules, with 90 biotech companies listing and raising HK$142.9 billion since 2018. Additionally, the Chapter 18C mechanism for special technology companies has facilitated listings for AI, robotics, and autonomous driving firms. In contrast, the A-share market experienced a broad decline, with over 4000 stocks falling, despite some gains in industrial gases and semiconductors. AI

OpenAI has announced its support for the EU Code of Practice on AI content transparency. This initiative aims to bolster Europe's efforts in creating a trustworthy AI ecosystem. The company's involvement will focus on advancing provenance standards and developing tools to help users identify AI-generated content. AI

Bangladesh and India have agreed to enhance border cooperation to address issues including migration and illegal crossings. The discussions between Border Guard Bangladesh and India's Border Security Force were described as positive, though Bangladesh has raised concerns about India's practice of pushing migrants back across the border without due process. This issue has become more contentious recently, with Bangladesh formally requesting an end to the practice. AI

Former US President Donald Trump announced plans to meet with top AI executives to discuss how these companies can give back to the public. He suggested that the government might hold stakes in these companies, with future profits benefiting American citizens. Trump believes this initiative will be well-received and make the public very wealthy. AI

Three major Chinese online food delivery platforms, Meituan, Taobao Flash Sale, and JD.com Waimai, have signed a self-discipline convention aimed at improving food safety and promoting high-quality development in the sector. This agreement, facilitated by the Guangdong Administration for Market Regulation, introduces a cross-platform blacklist system. This system will enable joint restrictions on merchants who commit serious violations, ensuring that a violation on one platform leads to restrictions across all participating platforms. AI

Argentina's President Javier Milei has proposed legislation to allow for "non-human corporations" that would be owned and operated entirely by AI agents. This initiative, part of a broader deregulation effort, aims to attract tech investment by removing the requirement for human oversight and accountability in corporate structures. However, critics like Yuval Noah Harari raise concerns about accountability and the potential for AI-run companies to manipulate systems for their own benefit, drawing parallels to historical corporate exploitation. AI