Across Southeast Asia and Hong Kong, finance teams are transitioning from ledger-keepers to strategic drivers of business resilience. The need for real-time cash visibility clashes with fragmented markets, pushing firms beyond digitisation. Technologies like automation and AI are viewed as essential, with 95% of regional tax and finance leaders prioritising data and AI tools to support innovation and predictive analytics.
Yet adoption is tempered by concerns over data integrity and the escalating threat of AI-enabled fraud. Singapore lost S$913 million to scams in 2025, recovering only S$140.5 million.
In the first half of 2026, Hong Kong recorded 20,613 overall deception and fraud cases, with total financial losses reaching HK$3.5 billion; investment scams cost victims roughly HK$1.65 billion, nearly half of all monetary damage from fraud during this period.
CFOs are prioritising robust verification controls, with 58% giving equal priority to payment speed and security, yet only 43% rate their ability to deliver both as strong.
The new mandate for the modern CFO
The Chief Financial Officer role has undergone a fundamental transformation. No longer confined to stewardship and reporting, today’s CFO is expected to architect AI value, shape technology strategy, and build organisations that can adapt as fast as the world changes.
According to Gartner research, navigating AI-led change while keeping organisations aligned in a volatile environment has become the biggest challenge for finance chiefs.
Layered on top is a question of confidence: just 36% of CFOs feel assured they can drive meaningful enterprise impact from AI, even as investment in the technology continues to climb.
As Karthik Manimozhi, global president of Eftsure, observes: “This is a very interesting topic because you cannot build AI on top of a very poor data foundation, which is where many companies find themselves here today. Master data integrity is a major issue across companies and enterprises.”
The cure, he argues, lies in unifying the enterprise’s financial vocabulary and creating a connected view of finance.
The data foundation: Building trust before AI
Proliferating disparate systems, siloed data, and a lack of collaboration between finance teams and the broader organisation compound the challenge of data integrity. “There is latency in systems, which creates timing issues, data accuracy issues, and so on. Then add to it the age-old problem of fraud, manual errors, inefficient manual processes,” Manimozhi explains. “Add to it the AI layer. That brings in a whole new level of complexity.”
The visibility gaps are alarming. A Kiteworks 2025 survey found that 46% of organisations unable to track third-party vendors miss breaches, multiplying enterprise risk. The shadow AI crisis is equally concerning: 100% of companies have AI-generated code, yet 81% of security teams lack visibility into its use.
Manimozhi points to a McKinsey 2026 reportindicating that 58% of AI incidents in companies involve tools not authorised within the enterprise. “This is a major issue. You need to get control.” The solution, according to Manimozhi, is a “control tower architecture for payment integrity and trust”—a framework that provides visibility across the technology stack. “Everything starts and ends with data integrity,” he emphasises.
Predictive cash flow: From visibility to simulation
In an environment of high economic and environmental volatility, CFOs must mature cash flow visibility from a real-time snapshot to a predictive, AI-driven forecast that actively models uncertainty. “AI to the rescue,” says Manimozhi, “because what AI can do really well, the right agents, is they can access your system of record, system of engagement, and pull the data, which was very hard to do before.”
The goal is to build a digital twin for the enterprise. “The right question isn’t, what is my cash position today? The right question is: what happens if currency drops 8% and two shipping lanes are closed? What does it do to my cash position then? It is more simulation-driven.”
Amidst divergent monetary policies and tariff uncertainty, finance teams are leveraging AI and automation to strengthen working capital velocity and accelerate receivables. “In a very unstable and volatile world, you’ve got to unlock every advantage you have,” Manimozhi states. “And the biggest advantage you might actually have is sitting in your balance sheet frozen.”
The Hackett Group’s 2025 Working Capital Survey reveals that generative AI can help unlock $1.7 trillion in excess working capital. “If companies improve their liquidity management and try to move even one percentile closer to their category leaders, they can unlock $761 billion that is sitting frozen on their balance sheet,” Manimozhi explains.
AI-driven automation is delivering tangible results. “The average high-performing AR teams move around 39 days, compared to the industry average of 50 to 54. That’s an 11-day improvement in your DSO,” Manimozhi notes. The key, he argues, is “getting that trust embedded in your infrastructure.”
Always-on treasury: The infrastructure imperative
For finance teams operating across multiple time zones and jurisdictions, the aspiration of always-on treasury remains elusive. “Honestly, I don’t think not much has happened yet. It’s not a technology problem. It is an architectural and a design problem for finance organisations,” Manimozhi asserts.
The TD Bank 2025-26 Treasury Survey confirms this view: 80% of treasury professionals are stuck with manual processes, disparate systems, and ageing systems, unable to collate real-time visibility into their cash position. Only 17% have accomplished real-time positioning across their systems.
Manimozhi identifies the solution: “It’s about real-time bank connectivity. It’s about using APIs, not batch processing. It’s about monitoring intraday liquidity and not waiting for end-of-day reconciliation.”
Project Nexus, the BIS Innovation Hub initiative connecting domestic instant payment systems across India, Malaysia, Philippines, Singapore, and Thailand, represents a significant step toward addressing this challenge.
Combating deepfake fraud
The sophistication of fraud tactics has escalated dramatically with the advent of generative AI. The case of Arup in Hong Kong, where a financial professional was tricked into transferring $25.6 million during a deepfake video conference, exemplifies the new threat landscape. As Manimozhi observes: “AI has made fraud cheap and truth expensive. The moment fraud became software; trust has to be embedded in your infrastructure.”
The scale of the problem is staggering. J.P. Morgan research indicates that humans correctly identify deepfake videos only 40% of the time. Veriff’s 2026 Fraud Industry Pulse Survey found that nearly 74% of fraud professionals reported a noticeable rise in online fraud, with 75% specifically noting an increase in AI-driven fraud attempts.
Manimozhi emphasises that traditional verification methods are inadequate. “For every $1 of fraud propagated, it costs you $5.75 in controls, brand damage, operations, recoup, and all that different stuff. We are in this very unfortunate situation where the enemy is sending drones, and we’re shooting them down with very expensive missiles. It’s not sustainable.”
The answer, he argues, is a payment integrity and trust layer that provides visibility across the tech stack. Manimozhi poses three critical questions for every payment: “Are you paying the right person or payee? Are you paying the right amount? And are you paying them with the right timing? And all three matter.”
Navigating fragmented payment rails and regulation
For finance teams operating across multiple jurisdictions, balancing the drive for automation with the reality of fragmented local payment rails and complex regulatory landscapes is a persistent challenge. “You have to acknowledge the fact that complexity and fragmentation exist,” Manimozhi states.
“Proper automation is the one that’s built for exactly that: complexity and fragmentation. That’s the one that is not likely to break or is likely to survive and adapt,” he explains. “You cannot build automation for uniformity because that’s not how the world works.”
The ideal response is establishing a control tower that abstracts complexity and delivers value. Manimozhi describes the layers: an intelligence layer that understands processes and systems; a semantic layer with a unified vocabulary; a process engine with business rules and thresholds; AI orchestration with controls over autonomous agents; and, most importantly, an auditability layer.
“Trust in modern enterprises has to be rendered continuously. You need to be able to explain every decision that was made.” Karthik Manimozhi
Integrating carbon data into treasury models
Sustainability has moved beyond a branding exercise and become a bottom-line issue. “The moment energy price volatility and carbon footprint of your supply chain and the cross-border tariffs became real issues on your P&L,” Manimozhi explains, “it is something that affects profitability, affects your cash flow.”
The EU Carbon Border Adjustment Mechanism (CBAM), effective 2026, represents a significant financial exposure for businesses trading with Europe. BloombergNEF reports that the EU carbon border tariff is reshaping industrial trade flows.
Manimozhi advises CFOs to “tackle it the same way they would tackle any other volatility. You need to build a simulation. You’ve got to put in the exposure to energy price fluctuations, weather-related impact on supply chain, geo-impact, cross-border tariff impact.” This requires building carbon exposure into supplier contracts and hedging strategies. “This is no longer a branding thing. It is actually a liquidity issue, liquidity risk with certain probability and materiality.”
Upskilling finance talent for the AI era
Despite increasing AI budgets, few organisations have reached advanced capability. Gartner predicts that only 20% of finance organisations will pivot all talent-related investments to advanced digital capabilities by 2028, but the target needs to be 50% by 2027. Gartner’s 2026 survey reveals that acquiring and developing AI and digital talent is CFOs’ top near-term challenge.
Manimozhi argues that finance leaders need to think differently about AI. “It is not a tool you throw into your organisation and try to have the organisation adapt to it.” He draws an analogy from the mobile revolution: “
Karthik Manimozhi
When people tried to move from desktops to mobile, they tried to recreate the desktop experience on the mobile, and it did not work so well. What they really needed to do was create a mobile native architecture. I think we’re in a similar position for artificial intelligence and finance organisations. You need to create an AI native finance organisation architecture.” Karthik Manimozhi
This requires understanding which processes to reform, choosing the right foundational AI tools, ensuring data integrity, and training the workforce. “The good news is most specialised finance teams are already small. It’s not about replacing people with tech. It’s about human in the loop, augmented intelligence—how do you use AI to amplify the impact of your people?”
The evidence supports this approach. BCG’s 2026 analysis found that companies at the leading edge of AI-first finance have improved the predictive power of their finance models by 50% or more, automated 90% of reporting, achieved 80% touchless invoicing, and freed up more than 30% of finance team capacity for higher-value advisory work.
As Manimozhi concludes: “If you want meaningful change management, go through that transformation—data integrity, financial integrity, reform your workflow, reform your workforce, and reap the benefits.” The 2026 finance function is being rebuilt around these principles, and the CFOs who lead this transformation will define their organisations’ future.
Click on the PodChats player to hear Manimozhi elaborate on how CFOs can drive cash resilience with visibility, velocity, verification.
How can CFOs build a centralised data and governance framework that ensures integrity, enables AI, and prevents third-party data leakage?
How can CFOs evolve cash flow visibility from near-real-time snapshots to predictive, AI-driven forecasts that actively model for uncertainty?
How can CFOs leverage AI and automation to accelerate receivables and strengthen working capital velocity, preventing liquidity from being trapped amid tariff and monetary volatility?
To what extent do current bank partnerships and technology infrastructure support an “always-on” treasury capable of managing trapped cash and intraday liquidity across time zones?
How can CFOs move beyond human verification to deploy cryptographic, AI-powered controls that authenticate identities and payments before authorisation, countering deepfake fraud?
How can regional finance teams balance automation ambitions against fragmented local payment rails and complex, evolving regulatory environments?
How should CFOs integrate real-time energy costs and supply chain carbon exposures into treasury and cash flow models as sustainability shifts from branding to bottom-line impact?
With AI budgets rising but capability lagging, what strategic framework can upskill finance talent to oversee autonomous AI systems effectively?
