(Bloomberg) -- Follow Bloomberg India on WhatsApp for exclusive content and analysis on what billionaires, businesses and markets are doing. Sign up here.
Southeast Asia is fast emerging as an investment hot spot for AI leaders like Nvidia Corp. and Microsoft Corp., which are plowing money into cloud services and data centers. But the region’s own young tech companies are failing to capitalize on the boom.
While the world’s biggest companies are set to splurge up to $60 billion over the next few years in Southeast Asia as its young populations embrace video streaming, online shopping and generative AI, little is flowing to the region’s startups that have artificial intelligence at their core. Investors are wary about betting on unproven entities, and the region has yet to show it can produce innovative firms that can scale significantly.
Venture investment in Southeast Asia’s young AI firms has amounted to just $1.7 billion so far this year, out of about $20 billion for the Asia-Pacific region as a whole, data from Preqin shows. Only 122 AI funding deals have taken place in Southeast Asia in 2024, versus the APAC total of 1,845.
The disconnect is raising doubts about the emerging region’s ability to build up its private sector and compete with the US and China, the world’s AI leaders. Venture investors’ skepticism toward Southeast Asia’s AI efforts is weighing on the growth potential of its entire homegrown tech sector.
Globally, investors are racing to tap the AI opportunity — but for now their focus is largely on the US and China. The US snatched $68.5 billion in AI funding in 2024, while China took up about $11 billion, Preqin data shows.
On the surface, Southeast Asia and its population of 675 million have what it takes: it counts over 2,000 AI startups, which is more than South Korea and almost as many as Japan and Germany, a report by tech advisory firm Access Partnership showed. Singapore, the region’s major business hub, ranks third in the Global AI Index, scoring high on indicators including the number of AI scientists per million people.
But the broader region, with countries such as the Philippines, Indonesia, Thailand and Malaysia, is culturally and economically varied, complicating efforts to rapidly scale products and services. That has led to a perennial question asked by investors: can local tech companies profitably compete on the global stage?
“The region’s diversity in language, culture, and infrastructure makes it harder to create large, unified datasets — something AI solutions traditionally rely on to scale,” said Jussi Salovaara, managing partner and co-founder of Singapore-based early stage VC Antler.
AI investors are looking at so-called foundation models that underpin various services, the software engineering required to train and refine them, as well as the hardware enabling it all, said Sang Han, a partner at East Ventures. “All that isn’t happening at scale in Southeast Asia,” he said.
The region’s entire venture capital industry is also suffering from a dearth of exits exacerbated by weak IPO markets, illustrating how difficult it is for Silicon Valley’s private capital-fueled model to catch on in developing markets. Private funding of companies in Southeast Asia is set to drop to its lowest level on record, research from Google, Temasek Holdings Pte and Bain & Co. showed, slowing sharply from pandemic highs as investors become more choosy and capital becomes more expensive.
Local governments aren’t sitting still. Virtually all have developed their own national AI frameworks, with Singapore also providing funding to industry startups through its investment vehicles.
More is needed, and the region’s governments must work together to create a coordinated plan, said Kelvin Lee, co-founder of investment platform Alta.
“Countries in Southeast Asia are focused on vastly different agendas: some on advancing high-tech sectors, others on improving basic infrastructure and living conditions,” he said. “This divergence makes it hard to prioritize moonshot innovation on a regional scale.”
Yet Southeast Asia’s potential can’t be ignored. While its AI industry is sputtering, its digital economy as a whole is growing in the double digits both by revenue and profit, according to a Google, Temasek and Bain study. It has a growing middle class with rising incomes, and expanding mobile and internet user bases. It’s also seen as being relatively shielded from geopolitical risks stemming from US-China tensions.
Southeast Asia’s AI opportunity may lie early in the value chain, including the collection and organization of big data, said Weisheng Neo, a partner at venture capital firm Qualgro. “That’s what can help us build core assets that will lead to a competitive advantage.”
Some of the region’s more successful AI startups have emerged this way. Singapore-based Patsnap Pte has invested in collecting, cleaning and structuring data that has become the backbone of AI models. Over the past 17 years, it built up large data sets spanning patent, chemical, drug and food databases, used by customers like NASA, Tesla Inc. and Walt Disney Co. Now, the SoftBank Group Corp.-backed company is using that data to train its own sector-specific large language models and has added AI tools like natural language processing.
Indonesia’s Alpha JWC, one of the venture capital firms bullish on the region, has teamed up with the Pijar Foundation to create a sandbox to help connect talent and budding AI startups to some of the country’s largest corporations.
“Through this program, we have greater visibility on the different pain points large corporations face in integrating AI into their workflows, and the talent that’s available to solve these problems,” said Jefrey Joe, a partner at Alpha JWC.
Efforts like that are fueling optimism in the local startup industry that there’s still an opportunity to catch the AI wave. But the push requires more cooperation between all industry stakeholders, Joe said.
“Capital can only take us so far,” he said. “It’s all about the ecosystem — we need the regulator, governments, buyers, suppliers, consumers to come together.”
©2024 Bloomberg L.P.