MDDI 演讲稿 · 2025-11-10
陈杰豪高级政务部长在 Compact AI 研讨会 2025 上的主旨演讲
Keynote Address by SMS Tan Kiat How at Compact AI Symposium 2025
要点
- • 5 年内新加坡科技就业增长 24%——从 2019 年 17.2 万人到 2024 年近 21.4 万人；科技毕业生起薪比平均高约 25%。
- • 2024 年 SME 的 AI 采用率从约 4% 三倍至 15%；大企业从 40% 升至 60%。约 6 成科技人才在非科技行业工作。
- • 「双语人才」（bilingual talent）是关键——既懂技术、又懂业务领域；纯写代码的角色最脆弱。
- • 三件要记住的事：①能否把 AI 应用到真实业务问题上；②懂得 AI 应用的「风险面」（业务、法律、声誉）——尤其在医疗、金融这类受监管行业；③地缘政治：稳定与协作能力本身就是溢价优势。
- • 「竞争不是你 vs AI，而是你 vs 那些更会用 AI 的人」。
完整译文（中文）
MDDI 英文原文译文 · 翻译日期：2026-05-02
晚上好，
新加坡职工总会专业人士及主管协会（NTUC-PME）总监 Bernard 兄，
各位女士、先生：
我很高兴今晚和大家在一起。
2. 在我开始之前，请举手——在座有谁从事科技岗位超过 20 年？超过 30 年的呢？
3. 我接触电脑差不多是 30 年前——这个世界已经变了不少。
4. 今天，借助云原生（cloud native）与 AI，许多事情的做法已经大不相同。
5. 我知道大家心里多少有些担忧——这对新加坡的科技行业与岗位意味着什么？我们的科技生态如何应对 AI、我们要如何为巨大的机会摆好位置？
6. 我开讲之前再做一次小调查——在场谁是「在科技公司里」做科技岗位的？又有多少人是「在金融、医疗、物流等非科技公司里」做科技岗位的？
7. 大约各占一半——这也大致就是整个科技体系的比例。
8. 今天我会分享几个视角——第一，政府在帮助新加坡充分利用 AI 这类技术上正在做什么；第二，我们如何与 NTUC 等许多伙伴合作，让我们的科技劳动力为 AI 做好准备；第三，我认为在这一空间向前走需要哪些关键能力与技能。
9. 先讲几个数据。过去 5 年，新加坡科技就业增长了 24%——从 2019 年 17.2 万名科技专业人士，到 2024 年接近 21.4 万。科技毕业生的起薪也持续具竞争力——比应届生平均高出约 25%。
10. 第二，我们的公司在更多地使用技术。IMDA 上月发布的《新加坡数字经济报告》显示——2024 年 SME 的 AI 采用率三倍——从约 4% 升至 15%；较大公司则从 40% 升至 60%。
11. 公司在认真思考如何使用 AI。这些都是自我报告——他们用得好不好是另一回事——但他们的渴望与意愿是真实的。
12. 这一点也呼应了刚才的现场调查——大约 6/10 的科技专业人士在非科技行业工作——物流、医疗、金融、酒店、零售等等。
13. 这就是大家要记住的三个统计数据。下面我接着讲。
14. 鉴于这种需求——公司在寻找的能力、专长、画像与心性都在演化。我们如何跟上市场？我们如何让自己保持相关、保持有价值、保持可雇佣？
15. 这就引到我接下来的内容。我跟新加坡很多雇主聊过——大公司、小公司、本地与外资科技公司。对我而言——医疗行业是我比较熟悉的——「医疗+科技转型」是我在卫生部的一项重点。
16. 我们如何用 AI？我们如何在所有医疗场景中——不只是医院里，而是从「人口健康」（上游）一路到「促成个人行为改变」（下游）——广泛而深入地使用技术？这要求我们在医疗中大量、密集地使用技术。
17. 让我提三点供大家在前进时参考。
18. 第一，科技专业人士需要怎样的能力？相信在场各位都知道——AI 是相当重要的责任。问题在于——它会以多快的速度颠覆我们的行业？能力需求会以多快的速度变化？它对岗位（不只是科技岗位）的冲击有多大？
19. 要发挥有用的角色——我们必须思考——你能把 AI 应用到哪里。这是当下大家在自问的最根本问题之一。
20. 现实是——巨额资金正在投向 AI 模型的开发。最终谁来为它买单？是企业——只有当它给业务或生活带来增量价值时，领导者才会买单。
21. 因此 AI 必须被用来解决问题、改善效率、降低成本——本质上是「做一些以前做不到的事」。对公司而言——这就是底线。对消费者而言——如果一个工具让你生活更便捷，你愿意为它付费。
22. 归根结底，未来会有大量努力是关于「如何把 AI 应用到真实世界的问题上」。真实世界的问题在哪里？在行业里。
23. 在医疗行业，我们在寻找能预测癌症概率、加快诊断的 AI 模型。在每一个行业里，我们都要记住——AI 能帮我们解决哪些客户愿意为之买单的问题。
24. 这就出现了「双语人才」（bilingual talent）的角色——懂技术、但更重要的是懂业务、能落地真实问题的工程师。
25. 在新加坡，相对其他地方我们位置不错——因为我们以「商业枢纽」闻名。我们在使用 AI 时把保障措施做到位，并缓释相关风险。
26. 我和伊利诺伊大学的几位校友聊过。一位先生说他很担心——无论你做的是软件编码还是业务流程外包——如果你只做「纯写代码」的事，你的角色会比较脆弱。
27. 让你有价值、避免被裁的，是「同时把技术与业务领域知识应用起来」的能力，而不是只在纯技术位置上工作。这就是第一个要记住的点。
28. 第二个要记住的点是——使用 AI 时理解风险的重要性。这本质上是商业决策——伴随业务风险、法律风险与声誉风险。这在医疗、金融等受监管行业尤为关键——这些公司必须维持强声誉，并在全球供应链中扮演关键角色。
29. 第三——我们处在一个「技术成为地缘政治竞逐空间」的时代。各国把 AI 视为重要的战略优势——这就是为什么你看到两大 AI 强国之间的对话不断展开。当国家之间出现影响经济或系统的扰动时，「稳定」与「能与他人协作的能力」本身就是溢价优势。
30. 我们已经开始有「双语 AI 人才」——懂技术、能把它应用到业务上，解决人们愿意为之付费的真实问题——让方案既可商业化、又可持续。
31. 我们于是能放就位「在不同维度上都被信任」的方案。我们运营的生态环境，能充分利用全球最佳实践（在能力与模型方面）——我相信这把我们置于一个非常有利的位置。
32. 最后一点——我们能做什么？我认为最重要的投资，是「投资自己」——投资雇主在新加坡及本地区所看重的技能与能力。你的技能会非常宝贵——因为能解决新问题的人并不多——价值就在那里。
33. 我们也要思考「颠覆」——竞争不是「你 vs AI」，而是「你 vs 在别处更会用 AI 的人」。在新加坡——我们有一个体系，在「与生态各方协作」的能力上可以与世界任何地方匹敌。
34. 作为一个体系——我们在这里支持个人的职业生涯。即使我们是一个小市场，我也相信我们能把这件事做好——支持那些希望在这里为世界开发方案的个人、公司、企业、创新者。
35. 我会以「审慎乐观」（cautiously optimistic）来收尾——我认为有重要因素站在我们这一边。
36. 趋势就在那里——但只要我们用更协调的方式协作——前方就有光明的未来。
37. 谢谢。
英文原文
MDDI 官网原始记录 · 抓取日期：2026-05-02
Good evening,
Brother Bernard, Director of NTUC-PME,
Ladies and gentlemen.
I’m very pleased to join all of you here today.
2. Before I start, just a show of hands – who has been in a tech role for more than 20 years? What about more than 30 years?
3. I started learning about computers almost 30 years ago, and the world has changed quite a fair bit.
4. Today, things are done much differently with technologies such as cloud native and AI.
5. I know at the back of our minds, there are some concerns about what this means for the tech industry and roles in Singapore, how is the tech ecosystem gearing up to think about AI and how are we going to position ourselves for great opportunities.
6. Maybe let’s do one more poll before I get into my remarks – who in the audience is in a tech role in a tech company? What about a tech role in a non-tech company, such as financial services, healthcare, logistics?
7. We have about half from each, and that is roughly the proportion of the entire technical system as well.
8. Today, I will be sharing some perspectives – firstly, on what the government is doing in terms of positioning Singapore on how to fully utilise technology like AI. Secondly, what are we doing to work together with our partners like NTUC and many others, to prepare our tech workforce for AI. And thirdly, what I think are some of the key competencies and skill sets that are needed as we move forward in this space.
9. Let me start out by sharing some statistics. Over the last five years, tech employment in Singapore has grown by 24%, from 172,000 tech professionals in 2019 to nearly 214,000 in 2024. Tech graduates also continue to earn competitive starting salaries – about 25% more than the average fresh graduates.
10. Secondly, our companies are using more tech. According to the Singapore Digital Economy Report, which was published last month by IMDA, shows that AI adoption among SMEs tripled in 2024, from around 4% to 15%. Larger companies saw an increase in AI adoption from 40% to 60%.
11. Companies are seriously thinking about how to use AI. These are all self-reported, whether they use it well is another matter – but they are keen, and there is a desire to use AI in their business.
12. And this shows in the poll I did earlier with the audience. About 6 of 10 of tech professionals work in non-tech sectors. These sectors are logistics, healthcare, financial services, hospitality, retail, etc.
13. Thus, those are the three important statistics that we should keep in mind, as I share the next part of my remarks.
14. Given the demand, the kinds of capabilities, competencies, profile, and temperament that companies are looking for, are evolving. How then do you keep abreast of the market? How do we keep ourselves relevant so that we can continue to be valuable and employable?
15. This brings me to the next part of my comments, I've been speaking to a number of employers across Singapore – large and small, local and foreign tech companies. And for me, I'm more familiar with the healthcare sector as one of my priorities in the Ministry of Health is to transform healthcare and technology.
16. How can we use AI? How do we use technology across all healthcare settings across the board – not just in the hospital, but upstream from population health, all the way down to nudging behavior changes for individuals? That requires us to use technology extensively and intensively in healthcare.
17. Let me highlight three points for us to think about as we move forward.
18. Firstly, what are the kind of competencies needed for tech professionals? I believe that all of you here know that AI is quite an important responsibility. The argument here is how soon will it disrupt our industries, how fast the demand will change in terms of capabilities, and what kind of impact it might make on jobs. Not just tech jobs, but across the board.
19. To play a useful role, we need to think about where you can apply AI. This is one of the most fundamental questions people asking themselves now.
20. The reality is that a lot of money is being poured into developing AI models. Who is going to pay for it? Ultimately, it will be the enterprises – leaders will pay for it only if it adds value in their business or to their lives.
21. Thus, AI needs to be used to solve a problem, improve efficiency, cut cost – essentially do something that they cannot do before. For companies, that's the bottom line. For consumers, if you're using a tool to make your life easier, you're willing to pay for it.
22. At the end of the day, there is going to be a lot of effort to look at how to apply AI to solve real world problems. And where are real world problems situated? They're in industries.
23. In healthcare, we are looking for AI models that can help predict the likelihood of cancer and provide a faster diagnosis. In every sector and industry, we keep in mind the problems that AI can help us to solve, and customers are willing to pay for it.
24. There is a role for bilingual talents – engineers who understand the technology, but most importantly, know how to solve a real problem that requires the person to understand the business and apply the solution.
25. In Singapore, we are in a very good position compared to others elsewhere because we are known as a business hub. We ensure that we put in place safeguards when using AI and mitigate the risks involved.
26. I had a conversation with some alumni of University of Illinois. One gentleman told me that he was worried because whether you are in software coding or business process outsourcing, if you are functioning as a pure coder, your role is vulnerable.
27. What makes you valuable and helps you avoid redundancy is the ability to apply both technology and business domain knowledge together, rather than working from a purely technical position. So that's the first point to keep in mind.
28. The second point to keep in mind is the importance of understanding risk when applying AI. It's a business decision that comes with business, legal risk, and reputational risks. This is especially critical in regulated sectors like healthcare and financial services, where companies need to maintain strong reputations and play crucial roles in the global supply chain.
29. Thirdly, we are in a position where technology has now become a contestation space in geopolitics. Countries look at AI as an important strategic advantage and that's where you see the conversations happening between the two major AI powers. When there are disruptions that affect countries' economies or systems, having stability and the ability for people to work together becomes a premium advantage.
30. We are starting to have bilingual AI talents who understand technology and can apply it to business, solving real problems that people are willing to pay for, making solutions both commercialisable and viable.
31. We can then put in place solutions that are trusted from different aspects. We operate in an ecosystem environment that makes full use the best practices available in terms of capabilities and models and I think that puts us in a very good position.
32. This brings me to the last point: what can we do? Well, I think the most important investment that all of us can do is to invest in ourselves, in terms of the skill sets and competencies that employers are looking for in Singapore and in the region. Your skill set will be very valuable because there aren’t many that can solve new problems, and that's where the value lies.
33. We also have to think about disruption. The competition isn't between us and AI, but between us and those who can use AI better elsewhere. In Singapore, we have a system here that can rival any in the world, in terms of our ability to work with different players in the ecosystem.
34. As a system, we are here to support individuals in their careers. I think we can do this well notwithstanding the fact that we are small market. We are able to support individuals, companies, enterprises, and innovators who want to be here to develop solutions for the world.
35. I’ll end off my remarks by being cautiously optimistic – I think there are important factors that are in our favour.
36. The trends are there, but I think that as long as we work together in a more coordinated manner, there is a bright future ahead of us.
37. Thank you.