#26 Asia AI Policy Monitor
APEC AI Initiative, AI & Hiring Discrimination in Australia, China Restricts GenAI for Children, Japanese Anime Embraces AI
Thanks for reading this month’s newsletter along with over 1,800 other AI policy professionals!
Do not hesitate to contact our editor if we missed any news on Asia’s AI policy at seth@apacgates.com.
Our second newsletter wrapping-up May shows the continued multidisciplinary approaches to AI regulation across Asia. In privacy law, South Korea’s PIPC continues its push to address AI and privacy harms with investigations of the AI Digital Textbook. In IP - the debate with copyright owners and AI tech companies continues in Japan, with the anime industry actively embracing the technology, but not without controversy. At a multilateral level, China pushes more deals with BRICS+ and global South countries, such as Brazil; while the EU partners with other advanced economies like Japan, and the US embraces its chip restrictions focused on China. In general governance issues, China is restricting AI use by children, while reviewing previous enforcement efforts by CAC. Australia is providing practical guidance for AI corporate transactions with model AI contract clauses.
In cybersecurity issues, thousands of Indonesians are recruited via fake job adverts and trafficked to fortified scam compounds on the Myanmar, Thailand and Cambodia borders. Australian research shows that predictive AI hiring systems can entrench unlawful bias by downgrading candidates for reasons as trivial as a headscarf or an ethnically distinctive name. Vietnam has meanwhile launched national AI ethics guidelines aimed at fostering a digitally ethical society and promoting responsible AI deployment.
In Fiji, the Minister for Labor is updating the country’s first National Occupational Health and Safety Policy to manage emerging digital and AI-related risks and safeguard worker welfare.
Four cybersecurity agencies—the US, New Zealand, Australia and the UK—have jointly issued guidance to raise awareness of AI data-security risks and provide best-practice measures across the AI lifecycle. Their recommendations span development, testing and deployment, outlining robust strategies for proactive risk mitigation. At online forums in May, experts warned that Southeast Asia’s heavy reliance on Chinese technology exposes the region’s AI infrastructure to critical vulnerabilities.
Privacy
South Korea’s PIPC investigated the AI Digital Textbook for privacy violations.
Regarding the legality and transparency of personal information processing: Corrective action recommendation (KERIS)
First, in the case of AIDT, there is no clear legal basis for personal information processing, so personal information is processed based on the consent of the data subject under the Protection Act and processing for contract performance.
In such cases, the personal information processing consent form, personal information processing policy, etc. must transparently describe the items of personal information being processed, the purpose, the retention period, etc. so that the data subject can understand them, but it was discovered that some of the information was omitted.
▸(Inadequate cases) Occurrence of omission in notification of items collected, used, and provided (contact information, consultation contents, etc.) during the civil complaint handling process
Second, a learning data storage (HUB) is built within the AIDT integrated portal (KERIS), where student learning content usage history data* (hereinafter referred to as the “national-level learning dataset”) provided by each developer is stored for statistical purposes (e.g., composing an integrated dashboard) or future AI-based learning analysis purposes.
Intellectual Property
The Japanese anime industry is embracing AI, despite copyright concerns.
Japanese artists are already worried about the developments. According to a 2023 survey by Arts Workers Japan, which collected data from nearly 27,000 respondents working in creative fields, 94% of Japanese artists have qualms about AI infringement of their copyrights…
And AI is already part of the mix. The first AI-powered anime, "Twins Hinahima," was launched in March this year. In 2023, a short anime clip named "Dog & The Boy" released by Netflix Japan also used AI-generated artwork to paint the backgrounds.
There is currently a labor crunch in the anime industry due to its unattractive working conditions. A 2024 report by the Nippon Anime and Film Culture Association showed that workers were overworked and underpaid, with hourly rates below the country's minimum wage being common.
Multilateral & Trade
The APEC AI Initiative was discussed at a recent meeting in Jeju, South Korea.
The initiative promotes a shared outlook, capacity building and investment in sustainable AI infrastructure.
Japan and the EU ink a deal on cooperation for quantum computing, including on AI.
A direct outcome of this initiative is the selection of the Q-NEKO project, a collaborative effort involving 16 European and Japanese institutions. Backed by €4 million (approximately $4.5 million) in EU funding, Q-NEKO will advance software, hardware, and hybrid computing architectures with an emphasis on quantum-enhanced machine learning and artificial intelligence
The US further restricts China, Hong Kong, Macau and Iran in chip sanctions around the national security threat of AI.
Specifically, BIS has determined that access to advanced computing integrated circuits (ICs) and commodities subject to the EAR for training AI models has the potential to enable military-intelligence and weapons of mass destruction (WMD) end uses in Country Group D:5 countries (including China) or Macau.
China invites the UK to its AI summit.
Shanghai will host the World AI Conference this summer, including “high level meetings on global AI governance,” Beijing’s ambassador to the U.K., Zheng Zeguang, said in a speech in at the Sino-UK Entrepreneur Forum in London. “We hope the U.K. government will send its senior representative.”
The invite comes amid mixed signals from the British government about its China policy — with a flurry of more hawkish moves in recent weeks coming after months of work to try and build bridges between the two countries.
China and Brazil ink a deal on digital economy including AI.
Article 2: Commits to supporting and encouraging companies from both nations to explore cooperation potential, actively participate in Brazil’s new industrial-policy projects.
Article 3: Invites firms to strengthen cooperation in building digital-economy infrastructure.
Article 4: Pledges support for deepening industry digitalization and a data-driven economy by fostering emerging-tech platforms (e-gaming, audiovisual content, streaming), leveraging interactive intelligent technologies—5G, IoT, AI, digital platforms, big data—and integrating these into advanced manufacturing, trade and logistics, transport, corporate finance, education, health and other sectors
Article 5: Encourages government agencies, universities, research centres and companies to strengthen communication and exchange.
6: Fully backs the China–Brazil Investment Cooperation Working Group (under the High-Level Coordination Committee’s Economic-Commercial Subcommission) to align strategies, policies, regulations and standards on the data economy, AI governance and digital platforms.
Governance
China’s Cyber Administration conducted an evaluation of a recent enforcement operation on platform recommendation algorithms.
In terms of improving the review of recommended content. Continuously improve the recommended content review mechanism, strengthen the push of positive energy content, and prevent the algorithm from recommending vulgar and bad information. WeChat Video Account improves the dual mechanisms of "friend recommendation" and "algorithm recommendation", continuously iterates and upgrades the identification and crackdown model, and strictly prohibits typical bad information such as vulgarity and vulgarity from entering the recommendation pool. Douyin innovatively launched a hot spot verification mechanism to prevent malicious dissemination behaviors such as posing for fraud, counterfeiting and hot spots, and patchwork and editing.
Australia’s Digital Transformation Agency published a guide on Model AI Clauses for buyers and sellers of AI services.
Some clauses are the same as or similar to clauses in other DTA contracts or model clauses, for example, the privacy clauses. Buyers should review and amend these clauses as necessary to ensure consistency and avoid repetition.
Before preparing an approach to market or draft contract, Buyers should consider Australia's AI Ethics Principles, AI in government policy, AI Assurance Framework and associated guidance.
China’s Cyber Administration published the latest batch of deep synthesis service providers.
Article 19 of the "Regulations on the Administration of Deep Synthesis of Internet Information Services" clearly stipulates that deep synthesis service providers with public opinion attributes or social mobilization capabilities shall perform the filing, change, and cancellation filing procedures in accordance with the "Regulations on the Administration of Algorithm Recommendation of Internet Information Services".
China is restricting genAI use by children.
Primary school students in China are prohibited from independently using artificial intelligence tools that generate open-ended content, while educators must ensure that AI can complement but not replace human-led teaching, according to a new guideline issued on Monday by the Ministry of Education.
Human Rights, Labor, Cybersecurity
The US, New Zealand, Australia and UK cybersecurity agencies issued guidance on AI data security.
The goals of this guidance are to: Raise awareness of the potential risks related to data security in the development, testing, and deployment of AI systems; Provide guidance and best practices for securing AI data across various stages of the AI lifecycle, with an in-depth description of the three aforementioned significant areas of data security risks; and Establish a strong foundation for data security in AI systems by promoting the adoption of robust data security measures and encouraging proactive risk mitigation strategies.
The EU and Japan will conduct meetings on the impact of AI on labor.
The 2025 EU–Japan Workshop will explore how Artificial Intelligence (AI) is reshaping the workplace and driving economic competitiveness. This hybrid event will bring together EU and Japanese stakeholders to discuss AI’s role in transforming job roles, promoting innovation, and raising ethical and policy challenges. It will also focus on how AI can support sustainable and inclusive growth in both regions.
Cybersecurity experts warn that Southeast Asia’s reliance on Chinese technology will create vulnerabilities, esp in the AI sector.
At a recent two-part online cybersecurity forum held on May 7 and May 15, experts highlighted how Chinese firms such as Huawei have entrenched themselves as major players in the region’s digital infrastructure – especially in telecommunications, cloud computing and emerging artificial intelligence applications.
Fiji’s Minister for Labor said he will update protections for workers in light of the digital and AI economy.
He told the World Day for Safety and Health at Work in Suva recently that with the assistance of the International Labour Organization (ILO), Fiji had developed its first National OHS Policy to manage emerging risks and ensure a safe working environment amid new technologies.
“As we embrace Al and digitalisation, prioritising worker safety and health is crucial,” Mr Kumar said.
Indonesians are targeted by scam operations and trafficked to scam centers around Myanmar, Thailand and Cambodia.
“Thousands of jobs, usually IT-related ones, are circulated on social media, but they are fake,” Anis Hidayah, commissioner of the National Commission on Human Rights in Jakarta, told Rest of World. “They generally target those who already have passports. They are recruited very quickly without predeparture training. Later, they can be trained within only two days and start working.”
Many of the scammers are young men and women from Southeast Asia, China, Ethiopia, India, and other developing countries. Hundreds of thousands of them sign up for what they believe are staid office jobs doing digital marketing and online sales that they find on social media. But they are instead taken to fortified scam compounds along the border of Myanmar, Cambodia, and Laos and in the Philippines, run by Chinese criminal syndicates.
Research published in Australia indicate that AI-supported hiring risks greater instances of unlawful bias and discrimination.
It is estimated that around 30 per cent of Australian organizations and 42 per cent of global companies use predictive 1 artificial intelligence (AI) systems2 in recruitment, with these figures set to grow significantly in the next five years. 3 These systems employ machine-learning algorithms 4and techniques, such as natural language processing (NLP),5 to classify, rank, and score job applicants. They promise time and cost savings for employers, improved quality of hires, and a superior candidate experience.6 But AI hiring systems (AHSs) may also enable, reinforce, and amplify discrimination against historically marginalized groups. They have been found to discriminate against applicants who wear a headscarf7 or have a Black-sounding name,8 and when the system is unable to accommodate requests for reasonable adjustments to enable access by people with disability.9 In the most well-known example, an AI system developed by Amazon learned to downgrade the applications of job seekers who used the word ‘women’s’ in their curricula vitae(CVs).
Vietnam launches for AI Ethics Guidelines.
Dr. Le Linh Long, Deputy Director of ABAII Academy, shared: “This initiative is more than just a training program—it is a strategic move toward building a digitally ethical society.
In the News & Analysis
This analysis looks into Japan’s AI Governance and “light touch” strategy.
Still, significant challenges remain. Despite its high-level commitment to global leadership in digital-driven sustainability, Japan continues to face internal hurdles that could hamper its digital ambitions. METI’s 2024 report identified several gaps — low executive engagement in AI strategy, insufficient digital and AI-related talent and a risk-averse corporate culture that slows innovation. Adoption of AI in core sectors like manufacturing and retail remains uneven, and many firms have yet to scale from pilot projects to enterprise-wide transformation.
In light of these obstacles, Japan’s move towards ‘light touch’ AI regulation and coherent policy coordination appears to be both a timely and strategic response to domestic imperatives and global trends. Like many other economies, Japan increasingly views digital transformation as a key driver of economic growth. Its attempts to assume a leading role in the global AI domain should be understood within this broader strategic framework. If successful, Japan’s experience could offer a blueprint for navigating the complex terrain of AI and digital disruption.
Taiwan’s President Lai announced the formation of a soverign wealth fund with a focus on AI investments.
The fund will "take advantage of the strengths of Taiwanese businesses and be led by the government while harnessing the power of the private sector to build a global presence and connect with major target markets in the AI era," Lai said, without elaboration.
This overview provides basics for AI governance in China, Japan, South Korea, India and Australia.
The Asia-Pacific region presents a diverse and fast-evolving regulatory landscape for AI. China sets the bar for assertive government control, enforcing strong compliance and penalties. Japan opts for a more innovation-friendly, voluntary approach, though it may pivot toward mandatory rules shortly. South Korea has passed a landmark AI law, with implementation set for 2026. Meanwhile, India and Australia are building the foundations for national frameworks, but rely for now on existing sectoral and privacy laws.
Advocacy
Malaysia’s Personal Data Protection Department is conducting a public consultation on automated decision making and profiling until May 14.
Philippines Privacy Regulators are collecting public comments on biometric data collection until May 30.
Saudi Arabia opened comment on the Global AI Hub law until May 14.
Pakistan has an open consultation on its draft National AI Policy ongoing.
The Asia AI Policy Monitor is the monthly newsletter for Digital Governance Asia, a non-profit organization with staff in Taipei and Seattle. If you are interested in contributing news, analysis, or participating in advocacy to promote Asia’s rights-promoting innovation in AI, please reach out to our secretariat staff at APAC GATES or Seth Hays at seth@apacgates.com.