My Cabinet colleague, Minister Chee Hong Tat
Mr Anthony Tan, Group CEO, Co-Founder of Grab
Mr Suthen Thomas, Chief Technology Officer of Grab
Distinguished Guests
1. Good morning. Thank you for inviting me to join you at the launch of Grab’s Artificial Intelligence Centre of Excellence.
2. Two years ago, we launched our second National AI Strategy to harness AI to uplift and empower our people and businesses.
a. We believed that AI would unlock a new frontier of economic growth, by driving innovation and productivity within our economy.
b. We therefore set out to encourage and support companies to set up AI Centres of Excellence (CoEs) to build new capabilities, develop solutions and drive value creation through AI.
c. Besides uplifting industry knowhow, these AI CoEs would also serve to uplift the capabilities of our workforce, and enable us to train and grow our pool of AI practitioners with the skillsets to create, implement and deploy AI systems and solutions.
3. Grab is well poised to realise the potential of an AI CoE.
a. From a ride-hailing app, Grab has grown into an “everyday superapp” offering food delivery, digital payments, logistics and financial services.
b. Its suite of products serve millions of users and support the livelihoods of drivers and merchants across Southeast Asia.
c. These daily transactions across its platform offer Grab a valuable trove of data to further enhance its value proposition for customers and businesses, through the use of AI and machine learning.
4. For consumers, AI can add value by enabling platforms like Grab to offer a more targeted, more dynamic and more secure customer experience.
a. By analysing past purchase or user patterns, AI-driven tools can generate recommendations and push promotions tailored to individual preferences.
b. In addition, AI-powered forecasting and optimisation can enhance demand-supply matching for services like ride-hailing and food delivery and reduce wait times for users.
c. With real-time analysis of transaction patterns, AI can also enhance fraud detection and prevention, especially for digital payments and financial services.
d. Through AI, consumers can benefit from greater personalisation, better precision and stronger protection.
5. Besides benefitting users, platforms like Grab can also leverage AI to empower businesses.
a. AI-enabled tools can help businesses analyse their performance and offer customised advice on how to grow their business.
i. For example, Grab’s Merchant AI Assistant uses merchant-specific operational and transactional data to provide merchants with business advice, such as how to boost their marketing and outreach or how to adjust their pricing strategies.
b. AI-enabled tools can also help merchants enhance their productivity and save costs, such as by handling customer service inquiries and feedback or managing inventory more efficiently and effectively.
c. By assessing income flows, transaction patterns and behavioral data, AI-driven credit models can also offer merchants with personalised credit products, so that they may have access to working capital in a prudent and responsible manner.
d. These AI-powered applications will be especially useful for small businesses, many of whom lack the resources or knowhow to implement AI solutions on their own.
6. By supporting companies like Grab to establish AI CoEs, we want to catalyse the development and deployment of AI-driven applications that can drive greater value for both people and businesses.
7. We will also help companies develop the capabilities to be able to administer AI solutions at scale. This will require that they have the right models, infrastructure and talent to be able to establish a robust AI ecosystem. Let me speak on each of these in turn.
8. AI models provide the foundational architecture to drive the intelligence and insights needed to power AI systems and applications. Through their AI CoEs, we will support companies to acquire or develop strong foundational models to support their AI ambitions.
a. Today, companies can already access a number of Large Language Models (LLMs) to train and run their workloads.
i. For example, Grab’s Voice Assistant for the visually impaired leverages OpenAI’s GPT4 models to convert speech to text, process these instructions, and convert text output into speech.
b. However, these LLMs may not be fully attuned to a company’s unique data and business needs.
c. In addition to supporting the development of end-user applications, we will support companies to develop the right models that can be tailored for their specific context.
i. In this regard, I am pleased to note that Grab is taking an ambitious step to develop its own foundational embeddings model in-house.
ii. This model will be uniquely trained on Grab’s real-world data, and will therefore be well tuned to its ecosystem.
iii. By being able to more accurately represent its drivers, merchants and consumers, Grab can provide much more precise recommendations to these various users and provide a richer user experience.
9. We will also ensure that companies continue to have access to compute resources for their AI workloads. This includes –
a. Progressively ramping up our national DC capacity, from about 1.4 GW currently to about 2 GW by 2030;
b. Establishing partnerships with major compute players, ranging from chipmakers to Cloud Service Providers; and
c. Working with our international counterparts to ensure that we have continued access to high-end AI chips.
10. Last but not least, as we transform our businesses to embrace AI, we must also invest in our people and develop AI-ready talent.
a. At the national level, we want to raise the number of AI practitioners to 15,000 over the next three to five years by expanding programmes such as the TechSkills Accelerator and AI Singapore’s AI Apprenticeship Programme.
b. Through company-based AI CoEs, these practitioners can further hone their skills through real-life use cases.
i. I am therefore glad to learn that Grab will create over 50 high-value AI jobs this year alone, in positions spanning areas such as Product, Engineering, Data, and Analytics.
ii. I hope this number will grow in the years ahead, as Grab continues to scale its AI investments.
c. In addition to AI practitioners, we need to build up a base of confident AI users that are ready to use AI-powered products and services.
i. Grab’s AI upskilling portal is a good example of how companies can uplift the AI capabilities across their workforce.
1. This portal provides a central platform for all Grab employees to strengthen their AI capabilities.
2. Through access to technical resources and project opportunities, this portal allows Grab to upskill its workforce, build practical experience in AI, and in time to come, allow its employees to pursue career pathways in AI-related fields.
d. I hope that through AI CoEs, more companies will build up their pool of AI practitioners and equip their workforce to be AI-ready.
11. Congratulations once again to Grab on the launch of their AI CoE. I am confident that this CoE will accelerate Grab’s AI innovation, enable it to create more value for peoples and businesses, and drive its growth in Singapore and the region.
12. Through attracting and anchoring companies like Grab to drive AI value creation here, we can continue to create good jobs and opportunities for our people.
13. Thank you.