Asia AI Policy Monitor
2024 Edition 1
Welcome to the first edition of the Asia AI Policy Monitor of 2024! We are dedicated to monitoring and analysis of AI-related law, policy and official guidance from governments, and civil society around Asia. If you would like to contribute to a future newsletter – please contact our editors here.
Which country to watch in 2024 on AI Policy?
South Korea will be the country in Asia to watch in 2024 leading AI discussion globally and regionally. Why?
South Korea to Host Key Summits
South Korea will host the AI Safety Summit this year, following the inaugural iteration of this event organized by the UK at Bletchley Park at the end of 2023.
The Bletchley Declaration focuses on risk-based analysis especially of frontier models, increasing private sector transparency, and building appropriate evaluation metrics for safety and research. The declaration was signed by 29 countries with 7 from Asia (Australia, China, Indonesia, Japan, Philippines, Korea, and Singapore).
Why it’s important:
We’ve noted previously Korea’s unique stance in Asia regarding digital rights – the only country in the region to have a Digital Bill of Rights, and ranks highly on openness and rights in the region.
At the same time, Korea’s Digital Strategy indicates a strong emphasis on AI and AI computing leadership. And given the country’s outsized dominance in memory chips – a good bet is that the country will take industry and innovation concerns seriously in balance with digital rights.
South Korea will also host the Responsible Use of AI in the Military Domain (REAIM) conference, which was organized at the Hague in early 2023. This resulted in high level political declaration endorsed by 49 countries of which only 4 are from Asia (Australia, Japan, South Korea, Singapore). The declaration included provisions around transparency, addressing unintended bias, and consistency with international law.
What we think:
On the security side, South Korea must address one of the largest cybersecurity risks from North Korea – ranked as the highest state sponsor of cyberattacks. AI is seen as a major risk to increasing the complexity to cyberattacks.
ASEAN to issue AI Governance and Ethics Guidelines
At the next ASEAN Digital Minister’s Meeting, the group of 10 countries will issue their AI Guidelines. Singapore is leading the meeting, and reports suggest that the guidelines will cover the following areas, creating an industry-friendly operating environment.
Malaysia’s AI Code of Ethics
Minister for Science Technology and Innovation Malaysia indicated that the country will issue their guidelines early this year. Previously, the government indicated they would move legislation in 3 areas to facilitate AI, including updates to public and private sector data protection rules and cybersecurity rules.
What we predict:
More country guidelines
We anticipate more countries in the region will issue these guidelines to augment AI National Strategies, in lieu of hard legislation in the vein of the EU’s AI Act. Many countries that indicated AI legislation was coming, ultimately backtracked as the potential economic impact became clear, and risks come better into focus (see the example of Taiwan, going one way, then the other on AI legislation.)
Asian countries’ guidelines will likely not focus on existential risk that consumes much of the discourse in Europe, but be focuses on the employment and economic opportunities and risks presented by the technology. Examining the AI strategies of the Philippines and Vietnam are a good example of where the region’s concerns are directed.
To Watch in 2024: Asia's Copyright Offices Issue Guidance
Litigation is on-going in the US and UK regarding LLM copyright infringement in their training data and output. Some copyright offices in the region have issued guidance such as Korea – which indicated that copyright does not apply to genAI materials unless there is human creative intervention; and also indicated that copyright owners should be compensated for use of their materials in training AI.
3 of the top 5 usage countries for ChatGPT are in Asia (India, Japan, Philippines). As users in Asia create more images, text and sound with genAI tools, these works will be and currently are producing economic value. The innovation community will need to better understand how to protect these works, and address how existing protections work in the system.
Deepfakes, fraud, elections and disinformation
🥸
Another area to watch will be how governments respond to the proliferation of deepfake disinformation – either in elections (of which there are many in Asia coming in 2024, including Bangladesh, Taiwan, India, Indonesia and Pakistan), or just for increasingly complex online fraud.
Following Taiwan's successful elections of last week - we recommend global watchdogs review how deepfake and AI was used in the election, and publish a report with recommendations for other democracies facing similar situations.
Check out this:
Articles about a deepfake CNN Philippines broadcast – apparently used to promote a company.
We saw in 2023, India’s Ministry of Electronic and Information Technology (MeitY) issue guidance about deepfakes, emphasizing that intermediaries must adhere to existing rules that implicate any social media postings.
What we think:
Countries will need to not just rely on the terms of use by platforms such as YouTube to prevent the unauthorized use of a person’s likeness through deepfake technology. Relying on trademark, copyright (in the CNN case above), trade dress, right to publicity (or personality rights) and other IP related legal areas will need to be bolstered over the coming years. We may see sui generis rights created to make take-downs of unauthorized deepfakes easier for victims, such as a US proposed Federal Anti-Impersonation Right (FAIR) Act as proposed by companies like Adobe.
We are starting a new series profiling AI policy and law leaders in the Asia-Pacific. Do not hesitate to nominate someone you are interested in learning more about, or believe is leading the discussion on AI policy in the Asia-Pacific region.
Asia AI Policy Leader Profile:
He Ruimin
Chief Artificial Intelligence Officer & Deputy Chief Digital Technology Officer
Government of Singapore, Singapore
This week we are looking at the United Nations Secretary-General’s AI Advisory Body Members representing the Asia-Pacific region. Of the 39 advisors appointed to the group, 9 are from Asian countries.
Dr. He is Singapore’s Chief Artificial Intelligence (AI) Officer, where he leads a multi-stakeholder effort to achieve Singapore's strategic AI objectives, including developing and implementing Singapore's national AI strategy. He is also concurrently the Singaporean government’s Deputy Chief Digital Technology Officer. Dr. He was previously the Chief Adviser to the CEO of Grab, the leading ride-share company in Southeast Asia, where he oversaw Grab’s economics, analytics, growth, and safety. He has also personally developed multiple revenue-generating software applications, taught at various universities including the Lee Kuan Yew School of Public Policy and Nazarbayev University, and published papers in journals such as the American Economic Review. Dr. He. has a BS in Electrical Science and Engineering, and a PhD in Economics, both from the Massachusetts Institute of Technology.
Mr. He is the only representative on the Advisory Body representing an Asian government directly. Singapore has also played a major role in Southeast Asia and the region on AI, building tools such as AI Verify, a practical toolkit for building responsible and trustworthy AI.
Along with the other members of the Advisory Body, he will be moving the UN towards completion the a final report to be published at the Summit of the Future in the summer of 2024. The report’s guiding principles will be around inclusivity, public interest in not creating harms by AI, data governance, multi-stakeholder and embracing international law, including the UN Charter and the Sustainable Development Goals.
Join the conversation!
This LinkedIn Group moderated by the staff at APAC GATES convenes a network of professionals interested in Asia's AI Policy and Governance space.
https://www.linkedin.com/groups/14353524/
