‘Not trying to solve world hunger’: GoTo takes pragmatic approach to AI

GoTo, the Indonesian technology group and parent company of Gojek, is narrowing its AI investments to applications that improve customer conversion or cut costs, ahead of a wider rollout in 2027, president director and group CEO Hans Patuwo said at the Tech in Asia Conference 2026 on Tuesday (Sep 15). AI now handles 35 per cent of GoTo's customer complaint tickets with satisfaction scores 20 per cent higher than the norm, and AI bots in its financial services business have lifted collection rates by three percentage points, while 10 to 15 more applications are in the pipeline. GoTo uses a mix of frontier, open-source and smaller models — including seven-billion and nine-billion-parameter models and an upcoming 27-billion-parameter model — partly because it earns revenue in Indonesian rupiah while paying overseas AI providers in US dollars.

‘Not trying to solve world hunger’: GoTo takes pragmatic approach to AI Gojek’s parent company is focusing on practical applications that can improve conversion or cut costs, ahead of a wider rollout in 2027 SINGAPORE After two years of experimenting with artificial intelligence, GoTo is narrowing its bets. The Indonesian technology group, which is the parent company of Gojek, is focusing its AI investments on applications that can improve customer conversion or lower costs, ahead of a wider rollout of AI-driven products and services in 2027. “We are not trying to solve world hunger,” said Hans Patuwo, GoTo’s president director and group CEO, as he made the case for using smaller, cheaper AI models where they can do the job. “AI should transform your business, not make you become an entirely different business,” he added. Patuwo was speaking on Tuesday Sep 15 on the first day of the Tech in Asia Conference 2026 held at the Sands Expo and Convention Centre, in a fireside chat moderated by Maria Li, former chief operating officer of Tech in Asia. GoTo began experimenting with AI in 2024, Patuwo said. By 2025, it had started deciding which capabilities it should build itself and where it made more sense to rely on external providers. “There is a period of experimentation where we let 1,000 flowers bloom, but that cannot carry on forever,” Patuwo said. The company eventually settled on two main measures for its AI projects: whether they improve user conversion or reduce the cost of serving customers. AI projects compete for resources in much the same way as any other product it might develop. Some are already producing results. AI now handles 35 per cent of GoTo’s customer complaint tickets, with satisfaction scores for these interactions coming in 20 per cent higher than the norm. AI bots used for collections in its financial services business have also helped lift collection rates by three percentage points, while another 10 to 15 applications are in the pipeline. The push comes as GoTo’s financial performance improves https://www.businesstimes.com.sg/international/asean/msci-drops-s-e-asian-heavyweights-sembcorp-goto-and-ayala-land-global-benchmarks . It posted a net profit of 252 billion rupiah US$14.3 million for the second quarter ended June, its second consecutive profitable quarter, against a 375 billion rupiah loss a year earlier. Net revenue rose 31 per cent to 5.7 trillion rupiah, while adjusted earnings before interest, taxes, depreciation and amortisation more than doubled to 1.01 trillion rupiah. Keeping AI costs in check Rather than relying entirely on the largest commercially available models, GoTo uses a mix of frontier models, open-source technology and smaller models trained further on Indonesian data. Patuwo said the company has used models with seven billion and nine billion parameters, and is launching a 27 billion-parameter model. Part of the calculation is financial. GoTo earns its revenue in Indonesian rupiah, while tokens consumed from many overseas AI providers are paid for in US dollars. “If you continue to consume in US dollars, that’s not going to be very sustainable,” Patuwo said. For new applications, GoTo may initially pay to use a frontier model. Once it better understands what the application requires, it can move to smaller models or train open-source models further using Indonesian data, including local language and cultural nuances. For many of the company’s applications, smaller models can provide sufficiently fast responses at a much lower cost. Patuwo gave the example of a production system processing 10 million transactions a day. If each transaction involves calls to an external AI model, those costs can become expensive very quickly. “We will never be the best, but we could be the best for our particular application,” he said. “We will never be the cheapest, but we could be the cheapest and the lowest latency for some of the services that we need.” GoTo is also cautious about handing an entire process to AI. Regulatory work and high-touch interactions, particularly those involving Gojek drivers, are among the areas where it is less likely to be an early adopter. Applications are instead introduced in stages, with GoTo generally stopping at around 70 to 80 per cent adoption initially rather than immediately moving to full automation. Rethinking hiring AI is also changing how the roughly 4,000-strong company approaches its workforce. GoTo has generally not been backfilling roles left vacant by departing employees for the better part of a year, Patuwo said. Managers can still hire, but he has set a higher bar. “My question to them is I’m not against your hiring. Go ahead, go hire as long as you can promise the candidate that we are offering them a career, not a job,” he said. Patuwo sees three ways where AI could affect jobs: existing roles could be “revolutionised” as workers become more productive; employees could be retrained for jobs that require greater human interaction; while some roles could ultimately be retrenched. Careful hiring and natural attrition could reduce the need for the last of these, he said. GoTo could eventually also sell some of the technology it is developing for itself. Patuwo said the group has historically been a “tech-enabled consumer company”, but could increasingly become a “consumer tech company”, with some technology built initially for its own needs to be “productised externally” later. Decoding Asia newsletter: your guide to navigating Asia in a new global order. Sign up here to get Decoding Asia newsletter. Delivered to your inbox. Free. Copyright SPH Media. All rights reserved.