Tech Week Singapore: It’s all about AI
From innovation in cloud services to data centers, AI dominated the customer conversations and discussions at the recent Tech Week Singapore.
The hype around AI showed no signs of slowing down as the technology remains a hot topic of discussion at Tech Week Singapore. The two-day expo saw technology vendors, partners and distributors from various sectors converge to discuss and share knowledge on some of the latest innovations in their industries.
While Tech Week Singapore featured vendors from cybersecurity, data centers, cloud service providers, e-commerce, big data and marketing technology, AI or generative AI to be more specific, was at the core of all discussions. Put simply, businesses are beginning to see more value in implementing AI to their organizations.
Conversations in cybersecurity focused on how AI in cybersecurity is enabling cybersecurity teams to not just automate tasks but have improved cybersecurity visibility as well as automated threat detection capabilities.
In the data center hall, vendors shared how the increased demand in AI is fueling the greater need of compute power from data centers. This includes improving data center output while ensuring data centers are able to meet sustainability requirements by having the right infrastructure built in, such as cooling technologies and efficient energy supply. Data center operators also shared how AI tools are being integrated into data center management, addressing concerns on skill shortages.
According to Andy Kiwanuka, Managing Director of CloserStill Media (Asia), the success of AI deployment in Singapore is a testament to the investment the government in putting in harnessing these new technologies.
“Our cybersecurity part of the seven co-located events at Tech Week Singapore has a lot of focus on harnessing AI in the cybersecurity market to wrap around the industry as it grows. We have been running these events for the last 10 years and knowledge sharing is very important for the development of industries. We’ve seen some fantastic leaps from collaboration and innovations that continue to be shared at Tech Week Singapore,” said Kiwanuka.
Kiwanuka also pointed out that the speed of development and innovations are at an unprecedented pace.
“It’s not just the tech giants of the world that are exploding the industry. There’s more investment coming into AI and we’re fortunate to have these conversations in Singapore. People are learning about where the industry is going. There are leaders in the industry that are setting the agenda,” added Kiwanuka.
The role of AI Singapore
For Laurence Liew, Director of AI Innovation at AI Singapore, while AI has been around for some time, the advancement in generative AI is definitely making a difference to organizations around the world. In Singapore, Liew mentioned that the companies working with them all have a generative AI deployment plan.
“In the last one year, we have helped about 35 to 40 companies in their generative AI deployment. These companies range from mid-size businesses to enterprises to government agencies. And all these companies need to have a deployment plan. If they can't convince me on their deployment plan, we will not approve their project. Hence, we have very high deployments for the projects that we undertake,” explained Liew.
Liew also highlighted that while there are a lot of enterprises working on trials and POCs (proof of concepts), AI Singapore will not be convinced. However, when these organizations do work with AI Singapore, they will need to commit and be prepared to go into deployment.
“Of the over 100 projects that we have worked on and deployed, there are zero concerns about regulation. Why? Because I think we are still in very early stages and unlike the EU AI Act, in Singapore, there’s no regulation on AI per se. However, when you deploy into certain regulated industries, the existing regulation in those industries comes into play. This is how the Singapore government is framing the whole concept and that’s how we work with our customers too,” said Liew.
For example, if a customer wants to build a robot advisor or fraud detection for insurance, Liew stated that AI Singapore can build the model for them. However, if the customer is from a regulated industry, they need to be sure their deployment adheres to all regulatory requirements.
Solving the cultural and talent dilemma
The increase in AI deployment also creates the demand for skilled professionals in the technology. Globally, many companies still feel that the lack of AI talents is hindering their AI deployment. However, Liew believes that Singapore may have already solved this problem through their current initiatives.
The AI apprenticeship program was created to meet AI Singapore’s requirements for AI professionals. Today, it has become a successful model in solving the AI talent gap in the country. The success of the program was even praised by Margrethe Vestager, the executive vice-president for the European Commission.
During a recent forum, Vestager suggested countries and organizations to study Singapore’s AI apprenticeship model because it not only allows businesses to get to use AI but also get experts who are embedded in the local community.
With over 400 graduates from the program, Liew stated that only 20% are from a computer science background. The rest are from a variety of sectors and fields of study but have shown a huge interest in developing AI. Hence, the program places them with their domain knowledge to work on AI use cases. Liew proudly pointed out that there is a 100% placement rate for all graduates. Put simply, this program directly helps deal with the talent crunch in AI that organizations are facing.
On concerns if AI would lead to a digital divide not just in the region but among Singaporeans as well, Liew highlighted the AI for Everyone program which was launched back in 2018. To get AI deployed successfully, Liew explained that everyone needs to be comfortable and be aware of what AI is. By improving AI literacy, Liew believes the implementation and deployment will be more easily managed.
“Even before ChatGPT, we knew we wanted to overcome some of the low resource language constraints. This is why we were building traditional NLP algorithms and collecting data partnering with out neighboring countries. We knew we needed to take care of the low resource language because the American and even the Chinese model will not be able to fit,” said Liew.
Liew pointed out that for some generative AI use cases, it's not just about the accuracy of the language but also the cultural nuances. And this is where the SEA-LION family of LLMs that is specifically pre-trained and instruct-turned for the Southeast Asian region comes into play.