Historically, large banks and financial services giants have concentrated their immense power on maintaining the status quo and enlarging their already significant market share.
Whatever technology they adopted was at best aimed at efficiency and further cost rationalisation to widen margins, and the gap between them and any incumbent. Core banking remained archaic, and in some major international and Malaysian financial services groups, that technology is in its same base form since the 1990s, if not earlier.
As technology became more available and democratised at the start of the 2000s, they went into “siege” mentality and denied that they required more technology as they already had three channels - so who would require more? They denied that their processes, their technology stack, was an issue that required greater attention, and sat back as profits surged until a number of them nearly broke the global financial system in 2008.
The financial technology sector, as a necessity, had to suffer a cycle of creative destruction. All their entreaties to the entrenched financial services companies were not entertained, and in frustration they started to build around the available technology. They found that much of finance was being implemented and had been built on architecture that was available at that time rather than what was now available to deploy.
However, as we navigate near the midpoint of 2026, the narrative has fundamentally shifted away from the combative "disruptor vs. incumbent" trope that defined the previous decade. The "First Wave" of fintech (2010–2024) was an era of unbundling — isolating specific banking services like payments, FX, or lending and optimising them through superior user experience, unbundling services and lower fees as a result of simplicity and technology, not just technology alone.
"Next Frontier" is an era of sovereign infrastructure and autonomous agents. We are no longer merely digitising paper-based processes; we are rewriting the underlying architecture of global value exchange.
By analysing the maturation of Real-World Asset (RWA) tokenisation, the legislative clarity of the 2025 GENIUS Act, and the rise of Agentic AI and the impending arrival of Quantum Computing this article explores how fintech has transitioned from a niche "disruptor" to the very operating system of global finance.
To appreciate the current situation, we must acknowledge the "Efficiency Paradox" of the first wave. Between 2010 and 2022, billions in capital flowed into technology-inspired neobanks, niche payment providers and SaaS startups with the promise of making traditional banks obsolete or at least taking significant market share and “eating their lunch” in the parlance of the times.
However, as Pham & Nguyen (2025) noted, fintech adoption by incumbents often led to increased operational margins and executive compensation rather than the predicted "disruption" of legacy players. In fact, when we look at AI use case pilots at the large FIs, what seems to be their focus? Again, it is efficiency, which means fewer people and more margin. But has this helped the consumer? Is the customer experience better than before?
The first wave taught us that while the "front-end" is easily disrupted, the "back-end" — defined by balance sheets, regulatory licensing, and institutional trust — is far more resilient. The era of "unbundling" reached its logical conclusion in 2023, giving way to the Great Rebundling.
Neobanks realised that customer acquisition costs (CAC) for single-product users were unsustainable, leading them to aggressively seek banking licences or "rent" them via Banking-as-a-Service (BaaS) to offer the full suite of high-yield savings and credit products.
Financial inclusion, another feature of the first wave and for many digital banking licences a requirement from central banks, saw mixed results. While digital credit expanded access in emerging markets, it also created "digital debt traps" in regions with insufficient consumer protection.
As we see in the Airlangga et al. (2025) review, the success of the next frontier depends on moving from "access to credit" to "access to sustainable wealth-building tools." In Malaysia the regulator has seen this and set up the Consumer Credit Oversight Board (“CCOB”) and begun licensing procedures for BNPL and other off balance sheet lenders thereafter.
If 2023–2024 was defined by the novelty of Generative AI (LLM and their logical extensions - chatbots), 2026 is defined by Agentic AI. The "Next Frontier" has moved beyond models that merely inform users to models that act on their behalf within defined guardrails.
We read today that Anthropic has delayed the release of “Mythos” because internal testing revealed the model possessed "dangerous" and "hacker-grade" cybersecurity capabilities that surpassed the company's existing safety thresholds - and yet we still speak of guardrails as a future feature for IT systems?
The ‘when’ aside, what are some guardrails we can institute?
Agents require access to internal APIs and databases to gain the maximum advantage from them, therefore technical guardrails are required to prevent "excessive agency" and data leakage. Some required steps without the details are defined here:
Operational Guardrails are needed. These focus on the "Human-in-the-Loop" (HITL) model, and ensuring that AI agents’ actions remain auditable.
Traditional finance was built on manual workflows. In 2026, we see the rise of the Agentic Mesh. Instead of a bank employee validating identity documents in one system and performing risk checks in another, an autonomous AI agent now ingests data, triggers AML (Anti-Money Laundering) screening, applies internal risk models, and executes the decision.
Research by Krungsri Research (2025) suggests that Agentic AI has improved KYC (Know Your Customer) productivity by nearly 20-fold.
These agents operate 24/7, shifting compliance from a "periodic review" model to an "always-on" monitoring system. For the consumer, this translates to "Self-Driving Wealth Management" — where an AI agent doesn't just suggest a portfolio rebalance but negotiates rates across multiple protocols and executes trades to optimise for real-time tax-loss harvesting.
As AI becomes the primary actor in financial transactions, the industry has had to abandon "bolt-on" security. As argued in MDPI (2026), the Next Frontier requires threat-centric architectures. We are seeing security integrated into the model weights themselves, preventing "hallucinated" transactions and ensuring that agents cannot be social-engineered into bypassing wire transfer limits.
The most profound structural shift in 2026 is the migration of Real-World Assets (RWAs) onto programmable ledgers. The speculative "crypto" volatility of the early 2020s has been replaced by the institutional stability of tokenised treasuries and private credit.
This was led by the Institutional Catalyst when by late 2025, BlackRock, and Apollo, to name a few, entered the fray, and tokenised RWAs (excluding stablecoins) surpassed US$50 billion in market capitalisation (InvestaX, 2025). The catalyst was the entry of Tier-1 asset managers.
BlackRock’s BUIDL fund and Apollo’s tokenised private credit funds proved that blockchain rails could offer T+0 (Atomic) Settlement, eliminating the "float" and reducing counterparty risk.
The value proposition is no longer "blockchain for blockchain's sake." It is about asset productivity. A tokenised Treasury bond in 2026 is not just a digital record; it is a "composable" unit of value that can be used as collateral across multiple decentralised and centralised venues simultaneously.
This "granularisation" of assets allows for the fractional ownership of previously illiquid assets, such as commercial real estate or fine art, on a scale never before seen.
For years, the US fintech sector operated in a "grey zone" of regulation by enforcement. This ended on July 18, 2025, with the signing of the Guiding and Establishing National Innovation for US Stablecoins (GENIUS) Act.
The GENIUS Act was a watershed moment for several reasons:
1. Institutional Exclusivity:
It limited the issuance of "payment stablecoins" to insured depository institutions and licensed non-bank entities under Federal Reserve oversight.
2. The Yield Prohibition:
In a move to separate "money" from "investments," the Act prohibits issuers from offering yields on payment stablecoins, effectively ending the era of algorithmic or "yield-bearing" stablecoins that led to the 2022 market crashes.
3. HQLA Status:
It clarified that dollar-pegged stablecoins could be treated as High-Quality Liquid Assets (HQLA), allowing banks to integrate them into their core liquidity management.
The macro-impact has been staggering. As IMF research (2026) demonstrates, demand for stablecoins now serves as a significant structural support for U.S. Treasury yields, as these digital dollars must be backed by "cash or cash-equivalent" reserves.
While the West consolidated around the GENIUS Act and MiCA (EU), the rest of the world has been building a parallel financial reality. The "Next Frontier" is not a single global highway but a series of interconnected, sovereign "liquidity corridors."
The graduation of Project mBridge from the BIS Innovation Hub in late 2024 signaled the end of the US-centric "First Wave" of global payments.
In 2026, we see the full-scale implementation of BRICS Pay — an independent alternative to SWIFT designed to facilitate intra-BRICS trade using local currencies and central bank digital currencies (CBDCs).
This fragmentation poses the greatest challenge to fintech firms: Interoperability. The winners of the next decade will be the "bridge builders" — firms that can provide compliant, real-time FX and settlement between the dollar-based "GENIUS" ecosystem and the emerging CBDC bridges of the East.
The first wave of fintech was obsessed with Gen Z and retail "neobanks." The Next Frontier has pivoted toward the Small to Medium Enterprise (SME) and Vertical SaaS.
In 2026, lending is no longer based on a static, six-month-old balance sheet. It is algorithmic and embedded. Whether it’s a construction management platform or a healthcare SaaS, the "fintech layer" is embedded directly into the business workflow.
Lenders now use real-time telemetry — inventory turnover, customer churn, and invoice velocity — to offer "Just-in-Time" credit. This shift has significantly narrowed the global SME lending gap, as software-driven data provides a more accurate risk profile than traditional credit scoring ever could.
The First Wave of fintech proved that technology could make banking better; the Next Frontier is proving that technology is making banking different. We have moved from the era of "Fintech as a niche" to "Finance as a technology."
The transition from the UX-focused 2010s to the infrastructure-focused 2020s has been marked by a return to institutionalism — but with a programmable twist.
The winners of the next five years will be those who master three core competencies:
The frontier is no longer a distant horizon. It is the code currently being written into the global, programmable ledger of 2026.
The frontier is no longer a distant horizon. It is the code currently being written into the global, programmable ledger of 2026.
Code being written by code and with an infinite loop leading to rQOPS>1m where rQOPS = Q x f!
(Sources: FIDE Forum Insights)
