Global AI News letter·Issue 14

On February 28, China's Cyberspace Administration released the "Guidelines for Filing Government Mobile Internet Applications (First Edition)" (hereafter "the Guidelines"). The document focuses on key aspects of filing management for government applications, providing clear instructions on filing methods, procedures, and required materials. It requires all entities responsible for launching government applications to strictly follow the Guidelines before deployment.

Link: https://www.cac.gov.cn/2026-02/28/c_1773925231290620.htm

2. The Cyberspace Administration of China and seven other departments have issued the 'Measures for Identifying Online Platform Service Providers with a Large Number of Minors and Significant Impact on the Minors Group'.

On February 28, the Cyberspace Administration of China, the National Press and Publication Administration, the Ministry of Education, the Ministry of Industry and Information Technology, the Ministry of Public Security, the Ministry of Culture and Tourism, the State Administration for Market Regulation, and the National Radio and Television Administration jointly issued the "Measures for Identifying Online Platform Service Providers with a Large Number of Minors and Significant Impact on Minors" (hereinafter referred to as the "Identification Measures"). The Identification Measures establish quantitative criteria for platforms with a large number of minors, comprehensively define platforms with significant impact on minors from six dimensions, standardize the identification process, self-assessment requirements, and dynamic adjustment mechanisms, require platforms meeting the standards to proactively apply for identification, and stipulate legal actions for non-compliance. These measures clarify platform responsibilities to strengthen online protection for minors.

Link: https://www.cac.gov.cn/2026-02/28/c_1774010730056867.htm

3. National Data Administration holds signing activity of industry high-quality data set chain master task book

On February 27, the National Data Administration hosted a signing ceremony in Beijing for the Task Book of Chain Leader Units for High-Quality Industry Data Sets. During the event, 72 leading organizations in the field formally signed the Chain Leader Task Book. Over the past year, the construction of high-quality industry data sets has achieved significant progress, with both scale and quality improving simultaneously. This advancement has effectively supported the digital and intelligent transformation of industries, establishing a favorable pattern where "wherever artificial intelligence advances, high-quality data set construction and application follow." Liu Liehong, Secretary of the Party Leadership Group and Director of the National Data Bureau, outlined six key requirements for development, covering expanded coverage, scenario-driven approaches, and technological support, emphasizing the need to convert data advantages into momentum for AI development. The National Data Bureau will strengthen coordination with relevant departments and entities to advance data set construction, thereby fostering high-quality AI development.

Link:https://www.nda.gov.cn/sjj/jgsz/jld/llh/llhldhd/0228/20260228171848897152605_pc.html

4. Beijing releases white paper on industrial development and governance of generative AI large models

On February 28,2026, the 4th Beijing Artificial Intelligence Industry Innovation and Development Conference, part of the Zhongguancun Forum series, opened at the National Convention Center Phase II. During the conference, the Beijing Municipal Cyberspace Administration released the "White Paper on the Development and Governance of Generative AI Large Models in Beijing," establishing a "safety maturity" indicator system focusing on research and training, filing management, and other aspects to promote the concretization and standardization of governance requirements. The China Electronic Information Industry Development Research Institute published the "Global AI Industry Trend Insight," analyzing the global industrial landscape and technological evolution. The Municipal Bureau of Economy and Information Technology provided a policy interpretation of the "Beijing Intelligent Glasses Industry High-Quality Development Action Plan." The conference also featured 16 sub-forums, focusing on key areas such as AI security, healthcare, and finance. Currently, Beijing is accelerating the formation of a development structure characterized by "foundation capabilities + governance system + application traction." As a key layout area, Mentougou District has established a domestically innovative AI computing power cluster, promoting the operationalization of the Beijing Algorithm Registration Service Center and the Mentougou Data Asset Service Center of the Beijing International Big Data Exchange.

Link: http://bj.people.com.cn/BIG5/n2/2026/0301/c14540-41511904.html

5. The General Office of the People's Government of Hainan Province has released the "Hainan Province' Artificial Intelligence Plus' Action Plan (2026-2028)".

On February 24, the Hainan Provincial Government unveiled the "Hainan Province 'Artificial Intelligence+' Action Plan (2026-2028)", outlining initiatives to establish demonstration zones for distinctive applications like the 'Artificial Intelligence+' low-altitude economy. The plan involves deploying a province-wide intelligent low-altitude infrastructure network, fostering deep integration of AI technologies with low-altitude industries, and prioritizing specialized sectors such as low-altitude logistics, marine monitoring, tourism, and emergency rescue. These efforts aim to cultivate a secure, efficient, and intelligent low-altitude economy as an emerging industry.

Link:https://www.hainan.gov.cn/hainan/szfbgtwj/202602/62a7dd9b7a9a43cd9873aaaf2d76e4c8.shtml?ddtab=true

6. Jiangsu Province Holds Special Meeting on Advancing 'Artificial Intelligence + Healthcare'

On February 24, the Jiangsu Provincial Government convened a special meeting on advancing 'Artificial Intelligence + Healthcare,' chaired by Governor Liu Xiaotao. The meeting thoroughly implemented relevant important discourses and the 'Artificial Intelligence +' strategic decisions and deployments of the Central Committee of the Communist Party of China and the State Council, aiming to promote the standardized application of artificial intelligence in the medical field, establish comprehensive intelligent services covering prevention, diagnosis, and treatment, and meet the public's health service needs.

Liu Xiaotao emphasized the need to integrate artificial intelligence into all aspects and processes of healthcare, focusing on five key areas: building high-quality datasets, establishing public healthcare data platforms, developing specialized large-scale models for clinical diseases and specialties, enhancing computing power centers, and driving scenario-based applications across multiple fields, to better serve public health. He stressed the importance of refining and implementing relevant development plans, advancing key tasks through a checklist approach, while mobilizing multi-stakeholder engagement, cultivating interdisciplinary talent, fostering industrial pilot zones, and creating demonstration projects.

Link: https://www.jiangsu.gov.cn/art/2026/2/25/art_92841_11732148.html

7. Beijing held a press conference on building a global digital economy benchmark city in 2026

On February 28, the Beijing Municipal Government Information Office hosted a press conference to announce the 2026 Global Digital Economy Benchmark City initiative. The conference revealed that by 2025, Beijing's digital economy will generate over 2.4 trillion yuan in added value, representing an 8.7% year-on-year growth and accounting for 46.4% of its GDP. This achievement will position the city as the second-ranked global digital economy benchmark city in the index.

The meeting outlined six key priorities for 2026: formulating and implementing the 15th Five-Year Plan for digital economy development; advancing technologies like large models and 6G to global leadership; targeting an AI industry scale exceeding 550 billion yuan through innovations such as AI glasses to dominate embodied intelligence; upgrading government big model platforms for enhanced urban governance; expanding intelligent computing capacity by over 70,000 P and refining the data factor market; and promoting the global implementation of more than 10 Beijing-based solutions.

Link: https://www.beijing.gov.cn/shipin/Interviewlive/1366.html

(II) Law Enforcement and Judicial Updates

1. Jiangsu Communications Administration reports APPs infringing user rights

On February 26, the Jiangsu Provincial Communications Administration identified 22 apps violating user rights, with violations including: unauthorized collection and use of personal data; excessive data collection beyond authorized scopes; forced, frequent, or excessive permission requests; frequent auto-launch and associated launches; inadequate disclosure of app information on distribution platforms; mandatory use of targeted push notifications; and difficulties in account cancellation.

Link:https://jsca.miit.gov.cn/gzcy/gggs/art/2026/art_712a0f32dbff41189aa4871bcc55fcb1.html

2. Jiangsu Cyberspace Administration Releases Typical Cases of Internet Law Enforcement in 2025

On February 28, the Cyberspace Administration of Jiangsu Province cracked down on a series of illegal activities, including web page tampering, data breaches, unlawful processing of personal information, and unauthorized deployment of new technologies or applications. Key cases include:

In a case of website tampering, a Xuzhou-based internet technology company, a Changzhou-based equipment manufacturing company, and a Zhenjiang-based technology company were ordered by local cyberspace administration authorities to rectify their actions. The companies were fined and issued warnings for failing to fulfill cybersecurity obligations, lacking effective technical safeguards, or failing to promptly address system vulnerabilities, which allowed their websites to be altered with illegal and harmful content.

In data security and personal information protection cases, a tech company in Nanjing was ordered to rectify and warned for collecting personal data beyond authorized scopes in its app, including unauthorized access to user location permissions under the guise of weather services. Two tech firms in Wuxi and Suzhou were penalized for system vulnerabilities causing data breaches and failing to implement legally required data protection measures like anonymization and encryption. A Taizhou-based e-commerce platform was fined for user data theft due to weak password vulnerabilities.

Furthermore, a unit in Huai 'an was issued a warning penalty for its inadequate cybersecurity management system, which led to the internal network being exploited as a channel for unauthorized data export. A WeChat mini-program operated by an information technology company in Zhenjiang was ordered to rectify its practices for collecting students' personal information without explicit notification, posing a risk of data leakage.

In the field of new technologies and applications, an advertising design studio in Xuzhou provided deep synthesis services such as AI voice-over and voice cloning, but failed to conduct safety assessments as required and did not fulfill its duty to provide warnings, resulting in a legal order for rectification.

Link: https://jswx.gov.cn/dongtai/gongzuo/202602/t20260228_1312187.shtml

(III)International Cooperation Updates

1. Shanghai takes the lead in realizing online cross-border digital identity authentication for foreign-funded enterprises

On March 1, the Shanghai Municipal Market Supervision Administration and the Municipal Data Bureau jointly announced the launch of a nationwide pioneering online service for the entire registration process of foreign-funded enterprises based on cross-border digital identity authentication. This application, leveraging the Shanghai-Xinjiang cross-border digital identity authentication as a breakthrough, utilizes distributed digital identity, verifiable credentials, and blockchain technology to achieve full online processing of cross-border enterprise registration.

For years, overseas investors have been unable to complete online registration procedures due to the lack of digital identity verification systems. To address this, Shanghai has established a cross-border digital trust service platform, with Singapore—a country with mature digital infrastructure—as the pioneer. This innovative model creates a digital trust bridge between Shanghai and Singapore, enabling Singaporean corporate investors to complete cross-border identity verification online through authorized signatories. This replaces traditional paper-based notarized documents, achieving "no-meeting approval" and "zero-travel residency registration," thereby significantly reducing institutional transaction costs.

Link:https://www.samr.gov.cn/xw/df/art/2026/art_263b66d46e5f41d9bfcc90d747823d54.html

II. International Governance Developments

(I) Policy and Legislative Updates

1. Connecticut Attorney General issues AI memo

On February 25, Connecticut's Attorney General released a memorandum titled "The Application of Existing Laws to Artificial Intelligence to Protect Connecticut Residents," designed to guide businesses, government agencies, and the public in understanding legal risks, safeguarding citizens' rights, and promoting responsible industry development when using AI technologies. The guidance emphasizes that the current legal framework already applies to various AI scenarios, and the development, deployment, and use of AI systems must comply with existing legal obligations, including privacy protection, anti-discrimination, contractual obligations, and consumer protection.

Link:https://portal.ct.gov/ag/press-releases/2026-press-releases/attorney-general-tong-releases-memorandum-on-artificial-intelligence

2. California Privacy Protection Agency supports the Whistleblower Protection and Privacy Act

On February 24, the California Privacy Protection Agency (CalPrivacy) and California State Assembly member Pilar Schiavo jointly introduced the Whistleblower Protection and Privacy Act, a new legislation designed to enhance individual protection when reporting privacy violations and improve the enforcement of privacy regulations.

The proposed legislation seeks to significantly enhance the California Consumer Privacy Act (CCPA) by establishing a comprehensive whistleblower incentive and anti-retaliation safeguard mechanism. CalPrivacy emphasizes that in today's data-driven economy, companies routinely conduct complex internal data collection and processing, with violations often embedded within business processes. Regulatory agencies find it challenging to detect such issues promptly through external investigations alone. The whistleblower system is therefore seen as a vital complement to uncovering illegal activities.

Link:https://privacy.ca.gov/2026/02/calprivacy-sponsors-whistleblower-protection-bill/

3. Italy's Ministry of Defense releases artificial intelligence defense strategy document

On February 26, Italy's Ministry of Defence unveiled a strategic document titled "Artificial Intelligence and Defence: The AI Defence Strategy" (IA e Difesa: Strategia della Difesa in materia di Intelligenza Artificiale). This initiative aims to equip the entire Ministry of Defence system with cutting-edge, secure, and resilient AI systems to ensure national security and make critical contributions to international stability amid rapidly evolving threats. The strategy primarily includes establishing the Office of Artificial Intelligence (UIA) and the Defence Artificial Intelligence Laboratory (LIAD), while outlining 15 strategic directives covering governance, AI operations, and training.

Link: https://www.difesa.it/assets/allegati/90197/ia_e_difesa_2026.pdf

4. National Institute of Standards and Technology updates generative AI evaluation program

On February 23, the National Institute of Standards and Technology (NIST) updated its generative AI evaluation program on its official website, detailing the implementation progress and milestone schedules for tasks including Text-2026.

The updates include revisions to the Evaluation Plan and phased timelines, with adjustments to model capability assessment methods, test data organization, and evaluation procedures. This project is a key technical initiative under NIST's initiative to develop generative AI that is measurable, comparable, and risk-identifiable, providing foundational tools for future AI governance, standardization, and compliance evaluation.

Link: https://ai-challenges.nist.gov/text-2026

5. The South Korean government will expand the application of artificial intelligence in the "gray area" and "public works"

On February 26, the South Korean government announced plans to expand artificial intelligence applications in "gray areas" and "public domain works." This initiative seeks to leverage AI technology for more efficient handling of ambiguous legal and policy issues in these gray areas, while enhancing the intelligent development and utilization of works owned or created by public institutions. Specific policy directions may include refining legal standards for AI training data usage, promoting the open access to public data, and fostering innovative AI applications in public services to boost administrative efficiency and the intelligent level of public services.

Link: https://www.korea.kr/news/policyNewsView.do?newsId=148960048

6. The U.S. State Department calls on the diplomatic system to oppose foreign data sovereignty legislation

Reuters reported on February 25 that the U.S. State Department issued an internal diplomatic cable instructing its diplomats abroad to actively oppose foreign-backed legislation on data sovereignty and data localization. Signed by the Secretary of State, the cable requires diplomatic personnel to voice concerns about measures restricting cross-border data flows during bilateral and multilateral engagements, while tracking and evaluating related legislative developments.

The telegraph noted that regulations restricting cross-border data flows could increase compliance costs for U.S. businesses, hinder the global deployment of AI and digital services, and potentially undermine open internet frameworks. The directive also emphasized promoting the Global Cross-Border Privacy Rules Forum (CBPR Forum) as a collaborative mechanism for cross-border data governance, replacing the current approach of data localization. This move reflects the U.S. diplomatic push to advance policies favoring free cross-border data flow in global data governance discussions.

Link:https://www.reuters.com/sustainability/boards-policy-regulation/us-orders-diplomats-fight-data-sovereignty-initiatives-2026-02-25/

7. Australian Office of the Cyber Security Commissioner launches study on impact assessment of social media ban on teenagers

According to a February 26 report by AAP News, Australia's eSafety Commissioner's Office is conducting a study on the impact of a social media ban targeting teenagers. The ban requires digital platforms to implement age verification and restricts users under 16 from creating accounts.

The study will evaluate the enforcement effectiveness of the ban and analyze the policy's impact on children's mental health and digital behavior habits by tracking the use of home device applications. Meanwhile, members of the Youth Committee will participate in the research framework design to incorporate minors' perspectives. This study is part of an assessment initiative for the implementation effects of the Online Safety Act 2021.

Link:https://aapnews.aap.com.au/news/social-media-ban-study-to-track-thousands-of-families

8. Turkey's Personal Data Protection Agency warns of QR code phishing risks

On February 26, Turkey's Personal Data Protection Agency released a risk advisory titled "QR Code Risks: Quishing Attacks," detailing the definition, implementation methods, detection techniques, and key considerations for combating this cyber threat. Quishing is a phishing tactic that exploits the ability of dynamic QR codes to redirect users to hidden URLs. Attackers may distribute fake QR codes in public areas like restaurants or posters, or send them via email or text messages, tricking users into scanning and redirecting them to malicious websites to steal login credentials, financial data, or install malware. To mitigate these risks, the document advises scanning users to check for physical tampering (e.g., covering or pasting) on QR codes, remain vigilant about codes containing emergency phrases or unknown sources, verify the legitimacy of redirected domains before scanning, avoid entering sensitive personal information on suspicious pages, and keep their devices updated with the latest security patches.

Link:https://www.kvkk.gov.tr/SharedFolderServer/CMSFiles/MTY5OWVmNzMzOWE4YWM.pdf

(II) Law Enforcement and Judicial Updates

1. The U.S. Federal Trade Commission issues policy statement on age verification enforcement

On February 25, the U.S. Federal Trade Commission (FTC) released its Policy Statement on the Children's Online Privacy Protection Rule (commonly known as the Children's Online Privacy Protection Rule).

The statement clarifies the regulatory boundaries for platforms employing age verification technologies, with key provisions including: 1) Exemption from enforcement. The FTC will not initiate enforcement proceedings under the Children's Online Privacy Protection Act (COPPA) if a website or online service provider collects, uses, or discloses personal information solely to determine user age, and does not use such information for any other commercial purposes without explicit parental consent. 2) Compliance requirements. Operators must ensure their data processing complies with the minimization principle and properly handle the data after completing age verification. 3) Policy objectives. By defining regulatory expectations, the statement encourages platforms to adopt age verification technologies to strengthen online protection for minors.

Link:https://www.ftc.gov/news-events/news/press-releases/2026/02/ftc-issues-coppa-policy-statement-incentivize-use-age-verification-technologies-protect-children

2. Singapore Personal Data Protection Board investigates multiple data security incidents

On February 26, Singapore's Personal Data Protection Commission (PDPC) issued a ruling and accepted rectification commitments from three institutions following recent ransomware attacks and system breaches. A B2B e-commerce service provider was found in violation of data protection obligations for failing to secure approximately 39,000 individuals' financial data due to security vulnerabilities, while the other three institutions received acceptance for their timely remediation and strengthened control measures. The PDPC also released practical data security guidelines for enterprises, outlining core protection measures including multi-factor authentication, patch management, and security monitoring.

Link:https://www.pdpc.gov.sg/news-and-events/announcements/2026/02/new-commissions-decision-and-undertakings-on-26-february-2026

3. South Korea's three major TV networks have jointly filed a lawsuit against OpenAI, accusing the company of using news content for ChatGPT training without authorization.

According to a Yonhap News Agency report on February 23, three major South Korean broadcasters—KBS, MBC, and SBS—jointly filed a lawsuit at Seoul Central District Court against OpenAI. They accused the AI company of using their news content for ChatGPT training without authorization, demanding damages and a cease-and-desist order to stop using unauthorized materials in the training process.

This marks the first legal action by three South Korean broadcasters against global AI companies. The Korea Broadcasting Association (KBA) emphasized that the news content constitutes core intellectual property accumulated through decades of media operations. The KBA stated that large tech corporations, leveraging their financial and technological dominance, have appropriated this knowledge for domestic commercial gain—an act that cannot be justified as 'innovation.' The lawsuit also addresses South Korea's data sovereignty concerns.

Link: https://cn.yna.co.kr/view/ACK20260223003100881?

4. Singapore penalizes SESAMi and its subsidiaries for non-compliance with data protection obligations

On February 26, Singapore's Personal Data Protection Commission (PDPC) issued a ruling, imposing an S$8,750 fine and corrective orders on SESAMi (Singapore) Private Limited for failing to implement adequate security measures to protect its personal data. Its subsidiary, Abecha Private Limited, was also ordered to rectify similar data protection violations, as the non-compliant software had triggered breaches of data protection obligations.

Link:https://www.pdpc.gov.sg/all-commissions-decisions/2026/02/breach-of-the-protection-obligation-by-sesami-singapore-pte-ltd-and-abecha-pte-ltd

5. Reddit fined £14.47 million for violating children's privacy

On February 24, the UK Information Commissioner's Office (ICO) announced a £14.47 million fine against social media platform Reddit for its serious breach of data protection laws and illegal processing of children's personal information.

The investigation reveals that despite Reddit's terms of service prohibiting children under 13 from using the platform, the site failed to implement reliable age verification mechanisms until July 2025. Relying solely on user self-declarations, the platform's system was highly vulnerable to bypassing. This allowed the illegal collection and use of personal information from numerous minors under 13, exposing them to inappropriate and harmful content. Additionally, Reddit did not conduct the required data protection impact assessment by January 2025, failing to adequately evaluate potential risks associated with processing children's data. ICO (Internet Children's Online) stated that Reddit's actions violated the UK Data Protection Act's special provisions for protecting children's personal information. The fine is part of ICO's ongoing efforts to safeguard children's online privacy, with continued monitoring of the platform's age verification measures in the future.

Link:https://ico.org.uk/about-the-ico/media-centre/news-and-blogs/2026/02/reddit-issued-with-1447m-fine-for-children-s-privacy-failures/

6. Poland's Data Protection Authority has fined a courier company approximately $3 million for GDPR violations.

On February 23, Poland's Office for Personal Data Protection (UODO) imposed an administrative fine exceeding 11 million Polish zlotys on DPD Polska, a courier company, for serious violations of the General Data Protection Regulation (GDPR).

The investigation revealed two major violations in the company's handling of personal information—including customer names, addresses, phone numbers, emails, and bank accounts—during parcel delivery. First, it failed to sign personal data processing agreements with external carriers (LNH), who handle address labels and related data during cargo handling. Second, it inadequately authorized employees to process data, as the automatically generated authorization documents through its IT system lacked employee names and signatures, violating data protection policies. UODO's chairman noted that while the company claimed its transportation operations did not involve data processing, the carriers' actual involvement in handling goods necessitated compliance with Article 28 of the GDPR. The company was ultimately fined 6.251 million zloty for failing to sign data processing agreements and 5.209 million zloty for insufficient organizational measures.

Link: https://uodo.gov.pl/pl/138/4075

(III)International Cooperation Updates

1.61 regulatory agencies worldwide sign Joint Statement on AI-generated Images and Privacy Protection

On February 23, under the coordination of the International Law Enforcement Cooperation Working Group at the Global Privacy Assembly (GPA), 61 privacy regulators from over 40 countries and regions jointly signed the "Joint Statement on AI-Generated Imagery and the Protection of Privacy." The statement outlines four fundamental principles: First, establishing robust protection mechanisms to prevent the misuse of personal information and eliminate the generation of involuntary private images and other harmful content, particularly those involving children. Second, ensuring full transparency regarding the functions, protective measures, acceptable usage scenarios, and consequences of misuse in AI systems to safeguard the public's right to know. Third, creating efficient and accessible mechanisms that allow individuals to request the removal of harmful content involving their personal information, with swift responses to such requests. Fourth, implementing enhanced protective measures to address specific risks targeting children, while providing clear and age-appropriate information to children, parents, guardians, and educators.

Link:https://www.edps.europa.eu/system/files/2026-02/23-02-26_joint-statement-ai-generated-imagery_en.pdf

2.Canada and South Korea Jointly State Will Strengthen Cooperation in Addressing AI Security Challenges

On February 25, during the Canada-South Korea Ministerial Meeting on Diplomacy and Defense (2+2), the foreign and defense ministers of both nations issued a joint statement. The statement underscored that both countries recognize the security, safeguards, and credibility of artificial intelligence (AI) as rapidly becoming critical elements for global stability. Both nations are committed to leading the development of responsible AI governance, ensuring that the creation, deployment, and use of emerging technologies promote security, safeguards, transparency, and human rights respect, while fostering AI-driven innovation that benefits all. The two countries pledged to deepen cooperation in AI policies and standards, responsible AI usage, and preventing the malicious or destabilizing use of advanced technologies, while maintaining regular dialogue.

Link:https://www.canada.ca/en/global-affairs/news/2026/02/joint-statement-canadarepublic-of-korea-foreign-and-defence-22-ministerial-meeting.html

3.India and Sweden establish "Sweden-India Technology and Artificial Intelligence Corridor"

On February 25, the Indian Artificial Intelligence delegation and the Swedish Ministry of Business Affairs signed a letter of intent to establish the "Sweden-India Technology and Artificial Intelligence Corridor" (SITAC). This platform aims to facilitate structured exchanges among government agencies, industry stakeholders, startups, and academic institutions from both countries. The collaboration will focus on six key areas: 1) organizing conferences, seminars, and thematic workshops; 2) enhancing exchanges between India and Sweden's AI ecosystems; 3) conducting field visits to innovation centers and excellence hubs; 4) fostering interactions among businesses, investors, researchers, and policymakers; 5) identifying opportunities for joint innovation platforms and investment corridors; and 6) promoting the bilateral deployment of AI solutions in priority sectors.

Link:https://www.pib.gov.in/PressReleasePage.aspx?PRID=2232803&lang=1&reg=3

4.India and Israel sign MOU on AI cooperation

From February 25 to 26, Indian Prime Minister Narendra Modi paid a state visit to Israel at the invitation of Israeli Prime Minister Benjamin Netanyahu. During the visit, the two nations signed a Memorandum of Understanding (MoU) on AI cooperation and another MoU on advancing educational development through AI. The two countries reaffirmed their commitment to collaboration in the field of artificial intelligence. Recognizing AI's strategic importance as a key driver of innovation, they agreed to deepen cooperation in AI talent and expertise. Both sides welcomed the development of a framework to facilitate connections between AI professionals, research institutions, and innovative enterprises, including promoting professional collaboration and participation in leading innovation initiatives across both countries.

Link: https://www.mea.gov.in/bilateral-documents.htm?dtl%2F40828

5.2026 Latin America Digital Summit to be held in Madrid, Spain

The 2nd 2026 Latin America Digital Summit was held in Madrid, Spain on February 26-27. Under the theme 'Where Latin America Meets the World to Shape the Future of Our Digital Society,' the summit brought together global representatives from government, business, regulators, academia, and civil society, establishing itself as the annual flagship forum for the digital ecosystem in Ibero-America.

The summit features keynote speeches, roundtable discussions, and workshops, focusing on cutting-edge digital topics such as AI, 5G, cloud computing, and cybersecurity. It aims to help Latin America establish leadership in these fields, build digital bridges between Latin America and the Middle East, the United States, and the EU, and promote collaboration in areas like artificial intelligence and smart cities. The event also connects global innovation hubs such as São Paulo and Dubai.

Link: https://digitalsummitlatam.com/

6.2026 Digital Platform Summit to be held in Brussels, Belgium

On February 25, the 2026 Digital Platforms Summit, hosted by the Centre for European Regulatory Studies (CERRE), took place in Brussels. The summit focused on discussions about the Digital Fairness Act and the Digital Markets Act's impact on reshaping online markets and governance. Margrethe Vestager, former Vice-President of the European Commission, engaged in a dialogue with CERRE's founder, sharing cutting-edge insights on the EU's digital regulatory framework and industry development.

The summit brought together EU officials, representatives from tech giants including Google, Apple, Tencent, and TikTok, along with leading experts from academia and consumer advocacy groups. Through keynote speeches, roundtable discussions, and thematic workshops, participants contributed insights to refine and implement the next phase of EU digital regulations.

Link: https://cerre.eu/events/digital-platforms-summit-2026/

(IV) Research Updates

1. North Carolina State University and the Institute of Certified Public Accountants jointly released "Executives' Perception of AI Opportunities and Risks"

On February 25, North Carolina State University and the Institute of Certified Public Accountants (ICPA) jointly released a report titled "Executive Perceptions of AI Opportunities and Risks: A Global Analysis".

The report compiles insights from 1,735 executives across eight regions and eight industries, highlighting five key observations: First, while AI adoption continues to grow, uneven readiness persists with significant variations in implementation models. Critical gaps in talent, technology, and governance are emerging, alongside escalating and rapidly evolving AI risks. Second, early adopters are taking the lead, though these organizations face heightened risk pressures alongside strategic momentum. Third, most organizations encounter barriers in AI readiness, with only 24-27% reporting adequate capability or regulatory preparedness – less than one-fifth of small organizations meeting essential requirements. Fourth, AI impacts vary by geography and industry: emerging markets demonstrate higher strategic adoption rates, while North American and European countries remain more cautious. The mining, professional/business services, and transportation sectors show strong momentum driven by automation and analytics, whereas construction, wholesale, and retail industries continue to lag due to fragmented operations and legacy systems. Fifth, AI risks are rapidly escalating, necessitating strengthened AI governance, model risk monitoring, and enterprise-wide risk management frameworks.

Link:https://erm.ncsu.edu/wp-content/uploads/sites/41/2026/02/Executive-Perceptions-of-AI-Opportunities-and-Risks-A-Global-Analysis-FINAL.pdf

(V) Industry Updates

1. Anthropic accuses three AI companies of illegally distilling its AI models

On February 23, U.S. AI firm Anthropic released a statement on its official website, accusing DeepSeek, Moonshot, and MiniMax of illegally extracting capabilities from its Claude AI model through distillation techniques to enhance their own models.

In its official statement, Anthropic revealed that the three companies in question had engaged in over 16 million interactions with the Claude model through approximately 24,000 fraudulent accounts, using model distillation techniques to steal the model's capabilities. Anthropic asserted that these companies employed detection-avoidance measures during interactions, with prompts differing in quantity, structure, and focus from legitimate interactions, constituting intentional extraction of model capabilities rather than lawful distillation. The company further emphasized that such illegal distillation amplifies risks within the model, circumvents U.S. export controls, and jeopardizes national security. Anthropic pledged to enhance monitoring of model interactions and implement protective measures. All three accused companies have since responded to the allegations on their official websites.

Link:https://www.anthropic.com/news/detecting-and-preventing-distillation-attacks

ꄴPrevious： null

ꄲNext： null