Enterprise AI is entering a less glamorous but more consequential phase: turning the documents, contracts and fragmented records inside large organizations into data that software can actually use. Singapore-based fileAI is betting on that infrastructure layer, securing strategic investment from SMBC Asia Rising Fund and Singtel Innov8 as it expands in Japan and launches a new unstructured-data mapping product.
fileAI Secures SMBC and Singtel Backing as It Targets Enterprise AI’s Data Problem
The enterprise AI market is increasingly moving beyond chatbots and copilots toward the underlying data infrastructure needed to make automation reliable. fileAI, the company behind the fileForge platform, is taking aim at that problem with new strategic investment from SMBC Asia Rising Fund, the corporate venture capital arm of SMBC, and Singtel Innov8, Singtel Group’s corporate venture capital arm.
The investment will support fileAI’s expansion into Japan and increase its focus on financial services, where enterprises routinely process large volumes of contracts, statements, compliance records and other unstructured documents.
At the same time, fileAI has introduced fileScout, an AI-native solution designed to map and understand unstructured enterprise data before it is processed by downstream AI systems.
The distinction matters because enterprise AI often runs into a problem before the model ever gets involved: organizations have enormous amounts of information, but much of it remains trapped in PDFs, scanned documents, contracts, emails and other formats that are difficult to structure and validate.
fileAI is positioning its platform as an intelligence layer between those sources and the enterprise systems that ultimately need the information.
Tackling the unstructured-data bottleneck
Large language models have made it considerably easier to extract information from documents, but deploying that capability across an enterprise presents a different set of challenges.
Organizations need to know where information came from, whether extracted data is accurate, how different records relate to one another and whether the resulting output can be audited. Processing costs also become significant when AI systems repeatedly ingest large volumes of unstructured information.
fileScout is designed to address the mapping stage of that process. According to fileAI, the technology can reduce token consumption when handling unstructured enterprise data, while its broader platform combines AI-driven data capture, validation, matching and reconciliation.
The resulting records can then be pushed into business systems as structured, verified data.
That puts fileAI in a part of the enterprise AI stack that is becoming increasingly important: data preparation and workflow automation.
The competitive landscape is broad. Hyperscalers including Microsoft, Google and Amazon provide document intelligence, AI infrastructure and data services, while enterprise software companies such as Salesforce, SAP and ServiceNow are embedding AI into business applications. Specialist vendors are also competing around intelligent document processing, data extraction and AI agents.
fileAI’s pitch is narrower. Rather than attempting to become a general-purpose enterprise AI platform, it is concentrating on complex file and data workflows where unstructured information needs to become operational data.
Japan becomes a strategic market
The new investment also strengthens fileAI’s expansion in Japan.
The company previously partnered with JRE Ventures, the corporate venture capital arm supporting the JR East Group, in June 2026. That relationship provided a foothold for applying AI agents to legacy contracts and operational documentation.
The latest funding is expected to support a local Japan team spanning sales, engineering and customer success.
Japan represents a potentially important market for enterprise AI vendors because large organizations often operate with substantial collections of legacy documents and established processes that cannot simply be replaced by modern cloud-native systems.
For fileAI, the opportunity is therefore less about introducing another generative AI interface and more about connecting existing enterprise information to newer AI-powered workflows.
Financial services offers a high-value test case
Banking is another major focus.
SMBC’s participation gives fileAI access to a large financial-services ecosystem while reinforcing the company’s strategy around regulated, high-volume enterprise workflows. Potential applications include extracting information from financial statements and loan covenants, supporting KYC and customer onboarding processes, reconciliation and regulatory reporting.
These are demanding environments for AI because errors can have operational, financial and regulatory consequences.
That makes governance particularly important. An enterprise AI system handling financial documents cannot simply produce a plausible answer. It needs traceability, validation and controls that allow organizations to determine how a result was generated and whether it can be trusted.
fileAI CEO Christian Schneider said the company wants to build an “operating layer” for enterprise AI around trusted data and production workflows. Singtel Innov8 Managing Director Boon Ping Chua similarly highlighted the need for reliable structured data as enterprises expand AI adoption.
SMBC Managing Director for AI Transformation Mayoran Rajendra said the group sees growing demand for technologies that can make unstructured information more accessible and improve operational efficiency and decision-making.
From AI experiments to production infrastructure
The investment arrives as enterprises increasingly confront the difference between experimenting with generative AI and deploying it across critical operations.
A model can summarize a contract in seconds. Building a system that processes millions of contracts, validates the extracted information, reconciles it with existing records and sends reliable results into core enterprise applications is a substantially harder engineering problem.
That is where fileAI is attempting to compete.
Its strategy also reflects a broader evolution in enterprise AI: AI agents are becoming less about conversation and more about completing controlled business processes. In that model, an agent might capture information from a document, check it against another source, identify discrepancies and route the result into a workflow without requiring an employee to perform every intermediate step.
For enterprises, the value proposition is consequently tied less to model novelty and more to reliability, integration and measurable workflow improvements.
fileAI’s challenge will be proving that its technology can deliver those benefits at enterprise scale while competing against much larger technology providers. Its backing from SMBC and Singtel gives the company stronger connections into major Asian corporate ecosystems, but sustained adoption will depend on whether its platform can turn unstructured information into dependable automation across highly regulated and complex environments.
The broader direction is clear: as AI becomes embedded deeper into enterprise operations, the companies solving the data layer underneath those systems could become just as important as the companies building the models themselves.
Market Landscape
Enterprise AI is increasingly shifting from generative AI applications toward data infrastructure, AI agents and workflow automation. The next stage of adoption depends heavily on whether organizations can make fragmented enterprise information usable by AI systems without compromising accuracy, security or governance.
The opportunity is attracting both hyperscalers and enterprise software vendors. Microsoft, Google and Amazon are building increasingly integrated AI and data stacks, while Salesforce, ServiceNow and other business software providers are embedding agents directly into enterprise workflows.
Specialist vendors such as fileAI are taking a more focused route, concentrating on the difficult conversion of unstructured information into structured business data.
For enterprise technology teams, the important comparison is not simply which AI model performs best. It is whether the platform can understand proprietary data, preserve context, validate outputs, integrate with existing systems and operate under enterprise governance requirements.
Financial services provides an especially demanding proving ground because document-heavy workflows intersect with compliance, auditability and operational risk. If fileAI can establish a strong position there, its approach could potentially extend into insurance, logistics, telecommunications, manufacturing and other document-intensive industries.
Top Insights
- fileAI has secured strategic investment from SMBC and Singtel Innov8, strengthening its Asian enterprise AI strategy and supporting expansion into Japan and financial services.
- fileScout targets unstructured enterprise data mapping, helping organizations reduce AI processing costs while preparing documents and files for downstream automation.
- The platform combines AI data capture, validation, matching and reconciliation, turning fragmented information into structured records suitable for enterprise systems and workflows.
- Japan and banking are becoming strategic growth markets, where legacy documentation, regulatory requirements and complex workflows create strong demand for governed enterprise AI.
- The announcement reflects a wider shift toward agentic enterprise AI, where intelligent systems execute multi-step operational processes rather than simply generating or summarizing content.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
