MDDI 演讲稿 · 2024-11-19
杨莉明部长在保诚全球 AI 实验室(新加坡)启动仪式上的演讲
Speech by Minister Josephine Teo at Launch of Prudential's Global AI Lab in Singapore
要点
- • 保诚（Prudential）在新加坡启动全球 AI 实验室——3 个月内已识别 100+ 实操用例（理赔裁定、客户咨询响应等）。
- • NAIS2.0 一年回顾：政府层面（公职人员自建个性化 AI 机器人、SCDF AI 分诊助理、学校的 Simpler 工具）；研究层面（NUS AI Institute、SIT-NVIDIA AI Centre、NTU 医学 AI 中心、$1.2 亿新元「AI for Science」、新加坡承办 ICLR 2025 与 AAAI 2026）；产业层面（Coca-Cola、Philips 加入 A*STAR 制造业 AI CoE，礼来、AvePoint 设立 AI CoE）。
- • 国际合作：与美国 CET 对话（含 AI 安全研究院网络）；与中国「数字政策对话」（重点是影响社会和谐的内容风险）；与英国 AI 安全 MOU；与卢旺达共同发布《小国 AI Playbook》。
- • 工具：IMDA《Model AI Governance Framework》+ Project Moonshot；CSA《保护 AI 系统指南与配套指南》——着力推动跨国互操作。
- • 保诚是 AI Verify Foundation 成员——共同推进全球负责任 AI 标准与最佳实践。
完整译文（中文）
MDDI 英文原文译文 · 翻译日期：2026-05-03
本文已从早期版本的网站迁移过来——格式可能有不一致之处。
Mrs Josephine Teo（数码发展及新闻部部长）在保诚（Prudential）新加坡全球 AI 实验室启动仪式上的主旨演讲
保诚集团首席执行官 Anil Wadhwani 先生，
各位嘉宾，
各位同事与朋友：
下午好。
1. 很高兴回到保诚。我记得大约 3 年前来访过——那时新加坡正非常艰难——还在新冠中——就业市场非常黯淡。我们都希望能想办法带来希望——「SG United Traineeship Programme」就是让年轻毕业生能起步事业的方式之一。我很高兴——保诚那时招的不少学员至今仍在你们身边——这一定意味着你们做得很好。这一次——我很高兴我们能讨论一件非常当下、且有突破性潜能的事——不只是对公司本身、对整个保险业也是。
2. 保诚今天在新加坡启动「全球 AI 实验室」——离总理黄循财发布更新版《国家 AI 战略》（NAIS 2.0）将近一年。
3. NAIS 2.0 提出了「卓越」（excellence）与「赋能」（empowerment）这一对目标——让新加坡在保险与金融等特定行业培育 AI 卓越的高峰，同时让最广基的个人、企业与社区能带着自信、判断力与信任去使用 AI。
4. 一年过去——我们在整个 AI 生态都看到大量进展——政府、研究、产业。
5. 在政府内部——AI 已经被部署到许多有意义的用例中：
a. 比如——今年公职人员一直在为自己创建个性化 AI 机器人——它们建立在一组定制文档之上，回答问题就像「拥有深厚机构知识的资深同事」。
b. 新加坡民防部队（SCDF）正在开发一款 AI 分诊助理——准确记录紧急呼叫的关键细节——从而提升响应速度与质量。
c. 在我们的学校——一款叫「Simpler」的 AI 工具——能改写句子、用契合学生阅读偏好水平的词——帮助他们更容易理解复杂主题。
d. 我们也在公共部门稳步建立 AI 人才池——配备专门 AI 团队，并提供大量的实验与跨部门分享机会。
6. 同样——研究生态也在多方面前进：
a. 新研究中心已经成立——包括 NUS AI Institute、SIT-NVIDIA AI 中心、NTU 医学 AI 中心。
b. 我们的学生在首届「国际 AI 奥林匹克」上夺得 4 枚金牌中的 2 枚。
c. 10 月——黄循财总理宣布投入 1.2 亿新元的「AI for Science」计划——促进跨学科协作，帮助研究者推动科学发现的边界。
d. 我们的研究者在顶级 AI 会议广泛发表论文。新加坡也即将承办两场重要的 R&D 领导者级 AI 大会——
i. 2025 年的「国际表征学习大会」（ICLR）——AI 专家社群里相当知名的活动；
ii. 2026 年的「人工智能促进协会大会」（AAAI，「Triple AI」）。
7. 除了政府与研究社区里发生的所有这一切——令我个人非常受鼓舞的是——我们的产业也在热情地把 AI 用起来——提升生产力、创造价值、变得更具竞争力。
a. 一种「AI 实验与创新」的文化在生长。Google 的「AI Trailblazers」等举措——已催化了早期采用者的行动。我们看到他们识别真实世界的挑战、可以用生成式 AI 来回应——构建原型——把原型推到生产环境。
b. 公司、初创、研究者走到一起——开发回应共性用例与问题的 AI 方案——目标是在区域层面规模化它们的成功。
c. 比如——A*STAR 设立了「制造业行业级 AI 卓越中心」——把研究专家与产业实务者聚到一起——开发从「产品设计与质量保证」到「工业自动化与预测性维护」等领域的 AI 方案。可口可乐（Coca-Cola）、飞利浦（Philips）等公司已签约加入这一努力。
d. 新加坡航空（SIA）、星展（DBS）等本地行业冠军在积极投资 AI 发展。更多公司——包括礼来（Lilly）、AvePoint——已经或正在新加坡设立 AI 卓越中心——开发新的 AI 加持产品与服务，并改造自身运营。
8. 我很高兴——保诚今天同样启动其 AI 卓越中心——这将给新加坡日益活力的 AI 生态再添一份。
9. 据我了解——这个 AI 实验室计划研究的问题——是「更好理解客户、为他们交付更好体验、并产生有意义的业务影响」。令人印象深刻的是——过去 3 个月里——这个实验室已识别 100 多个实操用例——包括以速度与共情来裁定理赔、回应客户咨询。
10. 作为「卓越中心」——保诚的成功——能成为金融行业 AI 价值创造的展示——不只是新加坡——是全球。这就是为什么我很高兴——你们在这里开发的 AI 方案——可以扩展到其他地方——影响你们在亚洲与非洲的运营、服务、产品。这与我们对新加坡 AI 生态的看法非常一致——既给本地业务、也给全球业务带来推动。
11. 我可以向各位保证——新加坡政府将继续以「使能性的政策、框架、沙盒」支持 AI 创新——同时——我们将扩大 AI 人才的广度与深度——支撑像你们这样的能力发展倡议。
12. 我们对 AI 的巨大承诺感到兴奋——毫无疑问。同样——我们对「负责任 AI 使用」的承诺也不应有疑——我们认真在确保 AI 系统按公平、透明、问责的方式被设计。我们相信——开发 AI 用例的企业必须竭力缓释偏见风险、保护敏感个人数据、维系客户信任。当公司负责任地使用 AI——他们不仅为客户与股东创造价值——也帮助公众建立对 AI 的信任。在这项技术被广泛部署的当下——这一点至关重要。
13. 在这件事上——我们积极推进国际伙伴关系，并尽可能贡献：
a. 与美国——我们有一个「关键与新兴技术」（CET）对话——让我们识别合作领域。AI 安全是其中之一。事实上——就在我们说话的同时——旧金山有一场「AI 安全研究院网络」会议——新加坡是这个网络中少数被代表的国家之一。
b. 与中国——我们有一个「数字政策对话」——开局不错。我们都关心如何管理 AI 的风险——尤其在「影响社会和谐与凝聚」的内容方面。
c. 除了美中——我们与 ASEAN 同行合作发布了《AI 治理与伦理指南》。我们最近也与英国签署了关于 AI 安全的新「合作备忘录」。我们也与卢旺达共同发布了《小国 AI Playbook》。
14. 与此同时——我们继续支持企业进行「负责任 AI 的开发与落地」。我们不能只「谈」负责任 AI——必须有方式去测试 AI 系统的「负责任程度」、并提供框架给用户「保障」。例子包括 IMDA 的《Model AI Governance Framework》与 Project Moonshot——以及 CSA 的《保护 AI 系统指南与配套指南》。这些资源为企业提供务实、易用的指引——覆盖最佳实践、风险管理与安全考量。我们鼓励企业善用它们。同时——我们也在与国际同行合作——让这些指南更具「互操作性」——让公司在跨境时不必每次都为「该满足哪些标准」而苦战。
15. 在这件事上——我很高兴——保诚作为「AI Verify Foundation」的成员——与我们共享这种 AI 安全的承诺。通过参与该基金会——保诚正在帮助推进「为负责任 AI 制定全球标准与最佳实践」的努力。
16. 结束之前——我期待保诚 AI 实验室带来的有意义贡献。祝各位事事顺遂！
17. 谢谢。
英文原文
MDDI 官网原始记录 · 抓取日期：2026-05-02
This article has been migrated from an earlier version of the site and may display formatting inconsistencies.
Keynote Address by Mrs Josephine Teo, Minister of Digital Development and Information, at Launch of Prudential’s Global AI Lab in Singapore
Mr Anil Wadhwani, CEO of Prudential plc
Distinguished Guests
Colleagues and Friends
Good afternoon.
1. It’s great to be back at Prudential. I recall visiting about three years ago. It was a very tough time for Singapore. We were still very much in the midst of Covid, and the job market was so bleak at that time. All of us wanted to find a way to bring hope, and the SG United Traineeship Programme was a way for us to allow young graduates to jumpstart their careers. I am glad that a good number of trainees that Prudential took in that time are still with you. It must mean that you all did a great job. I am delighted that this time round, we get to discuss something that is very current, and also has the potential to be groundbreaking – not just for the company, but also for the insurance sector at large.
2. Prudential’s launch of its Global AI Lab in Singapore today comes nearly a year after Prime Minister Lawrence Wong launched Singapore’s refreshed National AI Strategy, or NAIS2.0.
3. NAIS2.0 articulated the twin goals of excellence and empowerment – for Singapore to develop peaks of AI excellence in specific sectors, including the insurance and financial sectors, while enabling the broadest base of individuals, businesses and communities to use AI with confidence, discernment and trust.
4. One year on, we have seen much progress across the entire AI ecosystem: in Government, Research and Industry.
5. Within Government, AI has been deployed in many meaningful use cases:
a. For example, this year, officers have been creating their own personalised AI bots that draw on customised sets of documents and can respond to their questions almost like experienced colleagues with deep institutional knowledge.
b. The Singapore Civil Defence Force (SCDF) is developing an AI-powered triage assistant that helps to accurately record the critical details of an emergency call, thereby improving the speed and quality of response.
c. In our schools, an AI tool called “Simpler” can rephrase sentences and use words that match students’ preferred reading levels, helping them to more easily understand complex topics.
d. We are also steadily building up the AI talent pool among public agencies, with dedicated AI teams and numerous experimentation and cross-sharing opportunities.
6. Likewise, the research ecosystem has moved forward in several ways:
a. New research centres have been established. These include the NUS AI Institute, the SIT-NVIDIA Centre for AI and NTU’s Centre of AI for Medicine.
b. Our students won two of four gold medals awarded at the inaugural International Olympiad for AI.
c. In October, Prime Minister Wong announced a S$120million “AI for Science” programme to promote inter-disciplinary collaborations and help our researchers push the limits of scientific discovery.
d. Our researchers have published widely at top AI conferences. Soon, Singapore will also be hosting two major conferences for R&D leaders in AI. These are
i. the International Conference on Learning Representations (ICLR) in 2025 – a well-known event within the community of AI experts; and
ii. the Association for the Advancement of AI Conference (“Triple AI”) in 2026.
7. In addition to all that is happening within the Government and research communities, I’m personally very encouraged that our industries are also stepping up enthusiastically to make good use of AI, to raise productivity, create value, and be more competitive.
a. There is a growing culture of AI experimentation and innovation. Initiatives such as Google AI Trailblazers have helped to catalyse action by early adopters. We have seen them identify real-world challenges that can be addressed with generative AI, build prototypes and bring these prototypes to production.
b. Companies, start-ups and researchers are coming together to develop AI solutions that address common use-cases and problem statements – with the aim of scaling their successes across the region.
c. For example, A*Star set up the “AI Centre of Excellence for the Manufacturing Sector”. It brings together research experts and industry practitioners to develop AI-based solutions in areas ranging from product design and quality assurance, to industrial automation and predictive maintenance. Companies such as Coca Cola and Phillips have signed up to be part of this effort.
d. Local industry champions like SIA and DBS are actively investing in AI development. More companies including Lilly and AvePoint have set up or are setting up AI Centres of Excellence in Singapore, to develop new AI-enabled products and services, as well as transform their own operations.
8. I am happy that Prudential is similarly launching its Centre of Excellence in AI today. This will add to the increasingly vibrant AI ecosystem in Singapore.
9. I understand the AI lab plans to work on exciting problem statements that seek to understand customers better, deliver a better experience for them, and yield meaningful business impact. It is impressive that over the past 3 months, the AI lab has already identified over 100 practical use cases. These include adjudicating claims and answering customer queries with speed and empathy.
10. As a Centre of Excellence, Prudential’s success can be a showcase of AI value creation within the finance sector, not just in Singapore but worldwide. That is why I am glad that your AI solutions developed here may be scaled up elsewhere, to impact your operations, services and products across Asia and Africa. This aligns very much with how we see the AI ecosystem in Singapore, as a boost to businesses locally and globally.
11. I can assure you that the Singapore government will continue to support AI innovation with enabling policies, frameworks, and sandboxes. We will, at the same time grow the breadth and depth of AI talent to sustain capability development initiatives such as yours.
12. There is no question that we are excited about the immense promise of AI. Equally, there should be no doubt about our commitment to responsible AI use. It’s clear that we are serious about ensuring that AI systems are designed with fairness, transparency and accountability. We believe that businesses developing AI use cases must make every effort to mitigate against risks of bias, protect sensitive personal data, and retain customers’ trust. When companies use AI responsibly, they not only create value for customers and shareholders, they also help build public trust in AI. This is essential as the technology is being deployed much more widely.
13. In this regard, we actively pursue international partnerships and contribute where we can.
a. With the US, we have a dialogue on Critical and Emerging Technologies which has allowed us to identify areas of collaboration. AI safety is one of those areas. In fact, as we speak, in San Francisco, there’s a meeting of AI Safety Institutes as a network, and Singapore is one of the few countries that are represented at this network.
b. With China, we have a Digital Policy Dialogue that has gotten off to a good start. We are both interested in managing the risks of AI, especially when it comes to content that affects societal harmony and cohesion.
c. Besides the US and China, we worked with ASEAN colleagues on a Guide on AI Governance and Ethics. We also recently signed a new Memorandum of Cooperation with the United Kingdom on AI safety. With Rwanda, we jointly published the AI Playbook for Small States.
14. At the same time, we continue to support businesses in responsible AI development and implementation. We cannot just talk about responsible AI; there must be some way of testing the responsibility of AI systems and to provide frameworks to give assurance to users. Examples include the Infocomm Media Development Authority’s Model AI Governance Framework and Project Moonshot, as well as the Cyber Security Agency of Singapore’s Guidelines and Companion Guide on Securing AI system. These resources offer pragmatic, easy-to-use guidance for enterprises, covering best practices, risk management, and security considerations. We encourage businesses to make good use of them. At the same time, we are also working with our international counterparts to make these kinds of guides more interoperable, so that companies do not have to always struggle with knowing what standards to meet when they cross borders.
15. On this note, I am heartened that Prudential shares our commitment to AI safety, as it is a member of the AI Verify Foundation. By participating in the Foundation, Prudential is helping to advance efforts to develop global standards and best practices for responsible AI.
16. In closing, I look forward to seeing the meaningful contributions from Prudential’s AI Lab. I wish you every success!
17. Thank you.