Enterprise AI has spent the past two years trying to escape the demo room. For banks, insurers, manufacturers and telecom operators, the hard part is not producing a clever chatbot, but getting artificial intelligence to work reliably inside old, messy and highly regulated systems.
Singapore-based fileAI is building for that less glamorous, but more valuable, part of the market. The company has secured investment from SMBC Asia Rising Fund, the corporate venture capital fund linked to Japan’s Sumitomo Mitsui Banking Corporation, and Singtel Innov8, the venture arm of Singtel Group.
The size and terms of the investment were not disclosed.
Also Read: fileAI secures strategic investment from JR East Group’s venture arm to expand in Japan
The funding will support fileAI’s expansion in Japan, where large enterprises are under pressure to digitise legacy processes without compromising governance, auditability or compliance. It will also go towards strengthening the company’s financial services capabilities and the rollout of fileScout, its new product for mapping and using unstructured enterprise data.
Unstructured data refers to information that does not sit neatly inside databases, such as contracts, PDFs, scanned forms, emails, invoices, policy documents, onboarding files and other formats that still carry much of an organisation’s operational knowledge. For companies in sectors such as banking, insurance, logistics and healthcare, these files are often where bottlenecks begin.
fileAI’s core product, fileForge, uses AI to capture, validate, match and reconcile data from such documents, before turning it into structured, audit-ready records that can be fed into enterprise systems.
“AI will become an operating layer for every major enterprise, but that future cannot be built on fragmented data, unreliable outputs or endlessly expanding computing costs,” said Christian Schneider, CEO of fileAI. He said the company’s third-generation processing pipeline and fileScout are designed to help organisations convert complex unstructured data into “trusted intelligence and production-grade workflows”.
Japan becomes a strategic test bed
The investment follows fileAI’s June 2026 partnership with JRE Ventures, the corporate venture capital arm supporting the JR East Group. That collaboration laid the groundwork for fileAI’s Japan presence and focused on applying governed AI agents to legacy contracts and operational documents.
Japan is a logical market for this kind of enterprise AI. The country has some of the world’s largest banks, insurers, industrial groups and transport operators, many of which still manage heavy volumes of paperwork and semi-digital processes. At the same time, an ageing workforce and chronic labour shortages have made automation more urgent.
For Southeast Asian startups, Japan has long been an attractive but difficult market. Buyers tend to be demanding, sales cycles can be long, and trust matters deeply. Corporate venture investors can therefore play a larger role than simply providing capital. In fileAI’s case, SMBC Asia Rising Fund offers access to banking and regulated enterprise networks, while Singtel Innov8 brings links to telecoms, infrastructure and enterprise customers across Asia, Australia and Africa.
fileAI plans to build a local Japan team across sales, engineering and customer success. That is important because enterprise AI deployment is rarely a plug-and-play exercise. Companies need local support to adapt workflows, integrate with internal systems, and ensure AI outputs can be checked, explained and audited.
This is also where fileAI is trying to position itself: not as a general AI tool, but as an infrastructure layer for companies that need AI to behave predictably inside mission-critical work.
From AI pilots to production workflows
Across Southeast Asia, many large organisations have already moved past the question of whether to experiment with AI. The bigger question now is how to put it into production without creating new risks.
Generative AI models can extract, summarise and classify information, but enterprises often need more than a plausible answer. They need traceability: where the data came from, whether it was validated, who approved it, and how it changed downstream systems. In financial services, a mistake in customer onboarding, covenant extraction, regulatory reporting or reconciliation can create compliance exposure.
fileAI says its platform is built around data capture, preparation, governance and orchestration. In simple terms, that means it is trying to turn messy files into clean business records, while leaving an audit trail.
Also Read: fileAI’s US$14M Series A fuels expansion of AI-driven document automation
The launch of fileScout fits into this broader shift. According to the company, the product maps unstructured enterprise data and helps reduce token costs. Tokens are the small units of text processed by AI models; the more tokens a system has to read and analyse, the higher the computing cost tends to be. For enterprises with millions of documents, reducing that load can matter commercially.
This is especially relevant in Asia, where many firms are eager to adopt AI but remain cost-sensitive. A bank, insurer or logistics group may have decades of documents in different formats, languages and systems. Feeding all of that into large AI models without a disciplined data layer can quickly become expensive and difficult to govern.
A crowded global field
fileAI is not alone in chasing this market. Its rivals include automation and intelligent document processing companies such as UiPath, Automation Anywhere, ABBYY, Hyperscience and Rossum, as well as cloud-based document AI services from Microsoft, Google and Amazon Web Services. Large consulting firms and systems integrators also build bespoke automation layers for banks and other enterprises.
Where fileAI will need to differentiate is in deployment depth, governance and regional fit. Global platforms have scale and distribution, but Asian enterprises often require localisation across languages, document formats, compliance expectations and legacy systems. That gives regional players an opening, particularly if they can prove reliability in heavily regulated sectors.
fileAI says it has processed more than 1 billion files across finance, insurance, supply chain, healthcare and core business operations. Its customers include MS&AD, Toshiba, PwC, KPMG, Nippon Paint and Keppel.
For the company, the next stage is about converting that operational track record into a broader regional and global push. Japan appears to be a key part of that plan, both as a major enterprise market and as a proving ground for AI in complex environments.
Boon Ping Chua, Managing Director of Singtel Innov8, said enterprises increasingly need to turn “complex, unstructured information into clean, structured data that enterprises can trust and use at scale”.
Mayoran Rajendra, Managing Director of the AI Transformation Department at SMBC, said the bank sees demand for solutions that unlock value from large volumes of documents and unstructured data, adding that fileAI’s capabilities could support “data accessibility, operational efficiency, and decision-making” in the AI era.
Also Read: Japan is moving into Southeast Asia faster than the West, and most brands haven’t noticed yet
The investment also reflects a broader pattern in Southeast Asia’s AI ecosystem. Instead of competing directly with foundation model giants, more regional startups are building application and workflow layers around enterprise pain points. The bet is that the next wave of AI value will not come from flashy consumer tools, but from fixing the hidden plumbing of business operations.
For fileAI, that plumbing starts with the files most companies already have — and the expensive, manual work still required to make sense of them.
The post fileAI expands in Japan with new backing from SMBC and Singtel Innov8 appeared first on e27.