While the US and China continued to push the frontiers of infrastructure, Southeast Asia quietly emerged as the world’s launchpad for agentic AI’s application layer. This is where the technology is being tested, localized, and scaled to solve real-world business problems in some of the most complex and diverse markets on the planet.

Many of our 2025 podcast episodes captured this evolution, revealing a playbook for building the agentic future, built on three core pillars: the horizontal platform, the enterprise-ready foundation, and the global-first mindset.

The Southeast Asia Advantage: Real-World Problems at Scale

What makes Southeast Asia the ideal launchpad for agentic AI? The region’s inherent complexity and fragmentation are not bugs; they are features. They force founders to build for scale and adaptability from day one. Our guests in 2025 highlighted several real-world problems that are uniquely solvable at scale from this region:

• Cross-Border Compliance: Navigating the labyrinth of regulations across 11 different countries is a massive challenge. As Gani.ai co-founder Bintang Hidayanto noted, what works in a single market like the US won’t work here. This forces companies to build AI that can handle a multitude of legal frameworks, making it inherently global-ready.

“When you talk about compliance, the complexity gets amplified whenever you do business in this part of the world…You try to replicate it here, it’s very different.” [5]

• Hyper-Localization of Language: The region’s linguistic diversity, with hundreds of languages and dialects, is a forcing function for building more robust conversational AI. As Robin Li of WIZ.AI explained, success requires mastering narrow, local use cases first.

“The builders are successful because their AI products are always focusing on narrow use cases and try to get success in just those use cases.” [2]

• Financial Inclusion for the Unbanked: With a large unbanked population, Southeast Asia is a rich source of alternative data. As Nobel Prize-winning economist Dr. Michael Spence noted about Surfin’s approach, this data is the key to unlocking credit for millions.

“They basically said you can use the behavioral, social, et cetera, data to move past that.” [6]

• The Global Hub of Business Process Outsourcing (BPO): As fileAI CEO Christian Schneider pointed out in his NYSE interview, Asia is the world’s back office. Building an AI automation platform here provides direct access to the most complex workflows and the largest enterprise customers.

“Being based in Asia is for us a very good starting point because most of the world’s business processes are actually outsourced to Asia in general. So we’re using that base as a foundation for building a global company.” [11]

The Horizontal Platform: Building for Scale, Not Silos

The most powerful insight from 2025 is that the future of AI is not in siloed, vertical solutions, but in horizontal platforms that can be adapted to a multitude of use cases. As Clare Leighton, COO of fileAI, explained, this approach is fundamental to their global strategy:

“We have focused on building proprietary models and proprietary AI components that allow our users to construct these workflows…rather than developing very specific point solutions that operate in their lane but cannot scale out.” [1]

This horizontal approach allows companies to build a single, unified solution that can be deployed across different departments and industries, from finance and HR to procurement and customer service. It’s about creating a set of core AI components that can be combined and recombined to automate a wide range of workflows.

The Enterprise-Ready Foundation: Precision, Preparation, and People

But a horizontal platform is only as good as its foundation. 2025 was the year that companies learned that you can’t just plug in an AI and expect it to work. You need an enterprise-ready foundation that addresses the core challenges of data, compliance, and security. This foundation is built on three key elements: precision, preparation, and people.

Precision is about focusing on narrow, high-value use cases. For SMEs, this means delivering a clear and immediate ROI. Fluid CEO Trasy Lou Walsh captured this sentiment perfectly:

“Our vision is actually to build a team of AI agents that will automate processes that currently take finance teams hours or weeks to do, to really complete them in minutes.” [3]

Preparation is about doing the groundwork before you even think about the latest generative model. Carro COO Zi Yong Chua issued a stark warning against building “AI for AI’s sake”:

“We’re not trying to build AI for AI’s sake—we’re really trying to get business value out of it. It stems from my role because I’m in charge of operations and processes. It’s really about how we streamline our people and processes, using Gen AI to facilitate that.” [4]

Finally, it all comes down to people. The talent driving this revolution is as important as the technology itself. As former fileAI Product and Engineering VP Tim Prugar emphasized the need for deep integration with proprietary data and workflows:

“fileAI’s customers have a justifiably high bar for excellence and accuracy, and we endeavor to meet that. We don’t get in arguments with customers about whether 80% accuracy is good enough. We talk about our plan to get to 100% and deliver on that.” [7]

And in high-growth markets like Indonesia, developing local AI talent is a top priority. Thothathri Srinivasan, Flip’s Head of Engineering & AI, shared his approach:

“You need to go to the smaller [cities] like Bandung, in some of these other universities producing solid engineers. Especially in the context of AI. Because I think the engineers coming out now are able to use some of these tools better than engineers probably like me, who’ve been around for 15, 20 years.” [10]

The Global-First Mindset: From Day One to Global Scale

Finally, 2025 was the year that the most ambitious AI companies in Southeast Asia embraced a global-first mindset from day one. They are not just building for their local market; they are building for the world.

As fileAI CEO Christian Schneider explained in his interview at the New York Stock Exchange, this global ambition is a core part of their DNA:

“We have so much opportunity here. At the same time, we understand there’s a lot of investment going into this space, mostly driven by the US, and so of course we see that opportunity as well.” [11]

This global-first mindset is also reflected in how these companies are building their teams. They are tapping into a global talent pool, with teams spread across multiple countries. As Schneider noted:

“There is a war for AI talent here in the US specifically, and so we have a chance to also tap into resources in Asia and even in Europe.” [11]

This is also changing how venture capitalists approach the AI landscape. Kai Yong Kang of GenAI Fund noted the overwhelming B2B focus of the current AI startup ecosystem:

“So interestingly, and not so surprisingly, 92% of AI startups are B2B focused. Go-to-market with big and large corporates is actually something that they really need help with. When Gen AI took off, enterprises realized that they don’t move as fast as the startups and because this is so new and they fear being left out, there’s a lot of FOMO. That appetite of working with AI startups in a very early stage spiked up.” [12]

And it’s not just startups. As Robert Subbaraman, Nomura’s Global Head of Macro Research, pointed out, even a mature market like Japan is turning to AI to solve its labor shortage:

“Japan is really going to embrace AI and be a big adopter of it, which will boost productivity.” [13]

The Future is Agentic

The ultimate expression of this global-first, application-centric approach is the rise of agentic AI. Unlike traditional AI systems that respond to user prompts, agentic AI proactively identifies patterns, anticipates needs, and executes complex workflows autonomously. This shift from reactive to proactive intelligence is transforming how businesses operate across Southeast Asia.

As Akhil Nigam, CPO of Finmo, explained, the real breakthrough is not just having individual AI agents, but creating a system where multiple specialized agents collaborate seamlessly. It’s not about a user asking a question; it’s about the system proactively identifying opportunities and risks:

“The AR/AP agent, the eCommerce agent. And you have these agents with system-defined prompts. And they look at those, crawl the data, and then they start looking at, and that’s pattern detection that they look at. And then the most important piece is the agents talking to the agents.” [14]

This vision of interconnected agents is already delivering tangible results. Fluid CEO Trasy Lou Walsh highlighted the dramatic efficiency gains for finance teams:

“Our vision is actually to build a team of AI agents that will automate processes that currently take finance teams hours or weeks to do, to really complete them in minutes.” [3]

This level of automation is only possible because of the horizontal platforms being built. As Tim Prugar noted, the goal is to empower users to build their own agents and workflows, putting the power of customization in the hands of those who know their business best. The platform provides the components; the users construct the solutions.

Looking ahead to 2026, it’s clear that Southeast Asia’s role as the launchpad for agentic AI’s application layer will only continue to grow. The lessons of 2025—the horizontal platform, the enterprise-ready foundation, and the global-first mindset—have laid the groundwork for a new wave of innovation, one that is not just about building the future, but about applying it.

References

[1] “fileAI Founders CEO Christian Schneider and COO Clare Leighton on Building a Global AI-Powered Automation Engine” (https://review.insignia.vc/2025/12/23/fileai-nyse/) 

[2] “Driving conversational AI adoption from China to Singapore then the world with WIZ.AI Senior Director of AI Strategy and Partnerships Robin Li | Call 191” (https://review.insignia.vc/2025/09/25/wiz-ai-robin-li/) 

[3] “What it takes to drive AI agent adoption for SMEs in Southeast Asia with Fluid CEO Trasy Lou Walsh | Call 198” (https://review.insignia.vc/2025/11/27/fluid-ai/) 

[4] “Must listen call before you undergo Gen AI transformation | On Call with Carro COO Zi Yong Chua | Call 180” (https://review.insignia.vc/2025/04/16/zi-yong/) 

[5] “The Money for Compliance AI is in Cross-Border Adaptability | On Call with Gani.ai founders Bintang Hidayanto and Timur Nugroho | Call 187” (https://review.insignia.vc/2025/08/14/gani-ai-2/) 

[6] “Building truly global financial inclusion with AI | Surfin AI Fintech Forum Exclusive Roundtable | Call 181” (https://review.insignia.vc/2025/04/11/surfin-ai-fintech-forum/) 

[7] “Part 1. What enterprises truly need to build AI Agents | fileAI Head of Product and Engineering Tim Prugar | Call 188” (https://review.insignia.vc/2025/08/29/file-ai-tim-prugar/) 

[8] “Part 2. From wrestling coach to exited YC AI exec to fileAI Product & Engineering Head | Tim Prugar” (https://review.insignia.vc/2025/09/05/file-ai-tim-prugar-pt-2/) 

[9] “Part 1. Flip Indonesia Head of Engineering & AI on his journey from Groupon, Netflix, Pinterest to Investing | Call 189” (https://review.insignia.vc/2025/09/07/tho-flip-indonesia-part-1/)

[10] “Part 2. How to develop AI talent in Indonesia with Flip Indonesia Head of AI & Engineering Thothathri Srinivasan” (https://review.insignia.vc/2025/09/08/tho-flip-indonesia-part-2/) 

[11] “fileAI Founders CEO Christian Schneider and COO Clare Leighton on Building a Global AI-Powered Automation Engine” (https://review.insignia.vc/2025/12/23/fileai-nyse/) 

[12] “An Unconventional Venture Playbook for GenAI in Southeast Asia with Gen AI Fund Partner and Insignia Ventures Academy Cohort 3 Alum Kai Yong” (https://review.insignia.vc/2026/01/06/kai-yong/) 

[13] “Robert Subbaraman, Nomura Global Head of Macro Research, on Japan’s return to the investor spotlight” (https://review.insignia.vc/2025/12/16/nomura-robert-japan/) 

[14] “How Akhil Nigam, Finmo CPO And Co-Founder, Is Redefining Treasury Management With AI And Strategic Partnerships | Call 199” (https://review.insignia.vc/2025/12/02/finmo-akhil-nigam/)