By FSCL Director Riaz Patel
Why Dubai, Malaysia, Labuan and New Zealand may matter more to the next phase of AI-enabled financial services than their size suggests
There was something in the Global Fintech Fest 2026 discussion around artificial intelligence that I found more significant than the familiar debate about efficiency. The real question is no longer how AI can make financial institutions faster or cheaper, but what happens when intelligence itself becomes part of financial infrastructure.
If AI can make specialised knowledge cheaper, faster and more accessible, where will the next generation of AI-enabled financial services actually be built? My instinct is that the answer will not necessarily come only from London, New York, Singapore or Hong Kong. I would watch Dubai, Malaysia, Labuan and New Zealand precisely because each offers a different combination of technology, regulation, international connectivity, infrastructure and institutional trust.
The distinction matters because AI is increasingly becoming an ecosystem question. The financial centre of the future will depend not simply on banks and fintech companies, but on regulators, data, computing infrastructure, cybersecurity, legal frameworks, talent, capital and the ability to convert advanced technology into regulated and trusted financial activity.
Dubai is perhaps the most explicit experiment. The Dubai International Financial Centre has announced its ambition to become the world's first AI-native financial centre, with AI embedded across regulation, infrastructure and talent. Its proposed full-stack AI Campus is intended to bring together regulation, training, compute and physical AI, while DIFC also plans to provide financial firms with access to advanced AI tools.
What interests me is the underlying proposition. Dubai is not merely inviting banks to come and use AI. It is asking what a financial centre would look like if its legal, regulatory, technological and talent infrastructure were designed around AI from the beginning.
That could become a meaningful competitive advantage. The next generation of financial centres may be judged less by how many institutions they host and more by how effectively they connect institutions with technology, regulators, data and capital.
The challenge for Dubai will be execution. An AI-native financial centre cannot ultimately be judged by the ambition of its strategy. It will be judged when an AI system moves from analysing information to initiating an action, and the institution has to demonstrate who authorised that action, what information the system used, what controls applied and who remains accountable.
Malaysia offers a different proposition and its opportunity lies in connecting financial services with the rapidly expanding technology and physical infrastructure surrounding AI. Malaysia has become an important Southeast Asian data-centre location, particularly around Johor, with investment from major technology companies. The scale of that expansion is becoming significant enough to create its own infrastructure questions, including electricity demand, demonstrating that AI infrastructure is ultimately a physical economic proposition rather than something that exists only in the cloud.
For me, this creates an interesting financial possibility. AI-native finance needs computing power, data centres, connectivity, cybersecurity and payment infrastructure, but it also needs institutions capable of turning that capacity into regulated financial products and services. Malaysia therefore does not have to choose between being a technology hub and being a financial hub. Its more interesting opportunity may be to connect the two.
This is where Labuan becomes particularly relevant. Labuan does not need to compete with Dubai by attempting to become another global financial capital. Its more compelling role could be much more specialised: A cross-border financial environment in which AI can be applied to treasury, wealth structures, insurance, investment vehicles, Islamic finance and other international financial activities.
The opportunity, as I see it, is to make specialised expertise more accessible. A smaller international financial centre does not necessarily need enormous teams performing every legal, compliance, analytical, treasury and risk function if increasingly capable AI systems can help institutions access that expertise more efficiently. The competitive advantage could therefore lie in orchestration rather than scale.
Malaysia provides the broader technology and infrastructure environment. Labuan can potentially provide a specialised international-finance layer within that environment. If the two are connected intelligently, the combination could become considerably more interesting than either proposition considered in isolation.
New Zealand represents another model altogether. It is a smaller financial market, geographically distant from the major Asian centres and without the scale of India, Malaysia or the UAE. That is precisely why I find its approach worth watching. New Zealand's national AI strategy deliberately emphasises adoption and application rather than trying to compete in foundational AI development.
There is considerable realism in that approach. Not every country needs to build a frontier AI model, just as every financial centre does not need to become another Silicon Valley. There is economic value in becoming exceptionally good at applying available AI safely and intelligently to banking, insurance, payments, wealth management, compliance and financial advice.
New Zealand also introduces something that will become increasingly valuable as AI becomes more autonomous - institutional trust. An AI system operating in finance will eventually have to answer questions that go beyond whether it works. What was it permitted to access? What decision did it make? Under whose authority? Could the decision be reviewed? Who is responsible if the system is wrong?
These questions become much more consequential when AI moves from assisting people to acting on their behalf. That transition is already becoming visible. GFF 2026 itself placed agentic AI alongside tokenisation and quantum technologies as technologies capable of reshaping financial infrastructure. India is also moving towards increasingly sophisticated AI-enabled financial systems, while regulators are simultaneously placing greater emphasis on data responsibility, consumer protection and supervisory capability.
At GFF, RBI Governor Sanjay Malhotra stressed that data should be treated as a fiduciary responsibility, while SEBI has said it is moving towards adoption of a global supervisory toolkit for AI. This is important because the financial value of AI will ultimately depend on trust.
An AI system that assists a compliance officer is one thing. An AI agent capable of initiating a payment, reallocating a portfolio or making a financial recommendation is something else. Once machines begin acting with financial authority, identity, permissions, auditability and liability become part of the technology itself. That could become one of the most important competitive distinctions between financial jurisdictions.
Dubai has the ambition to build around this future. Malaysia has the physical and technological infrastructure opportunity. Labuan has a specialised cross-border financial proposition. New Zealand offers a useful model of adoption accompanied by institutional trust and responsible governance.
I would therefore resist the temptation to view these jurisdictions as competitors fighting for the same market. They could instead become complementary components of a future financial architecture.
A financial institution could potentially use Dubai for international capital and market access, Malaysia for technology and regional connectivity, Labuan for specialised cross-border structures, and New Zealand for particular governance, innovation or market applications. The future financial centre may consequently be less a single location than a network of specialised jurisdictions.
That is also where I return to the idea of headroom. AI can create headroom by reducing the time spent on routine analysis, documentation, reconciliation, coding and other knowledge-intensive activities. But headroom is not itself a competitive advantage. It becomes one only when institutions use the additional capacity to do things that were previously too expensive, too complicated or too specialised to undertake.
A financial centre that uses AI merely to reduce operating costs may become more efficient. A financial centre that uses AI to make previously uneconomic financial services possible could become structurally more competitive.
There is, of course, a significant risk in this transition. AI is creating new financial-stability questions around concentration in technology providers, enormous investment in computing infrastructure and the possibility that financial institutions become dependent on a relatively small number of external technology platforms. The more deeply AI becomes embedded in finance, the more regulators will have to understand not only how banks use AI but how financial markets themselves become exposed to the infrastructure and companies powering it.
That is why I believe jurisdictional diversity matters. Malaysia can test the intersection of manufacturing, infrastructure and finance. Labuan can explore specialised cross-border financial applications. Dubai can push the AI-native financial-centre proposition. New Zealand can demonstrate how advanced technology can be adopted without sacrificing institutional trust.
And there is a broader lesson for financial centres watching this transition from the sidelines. I would not measure success by the number of AI funds announced, AI districts established or memoranda of understanding signed. Those may be useful signals, but they are not the real test, which is much harder.
Can a financial institution take a genuinely difficult problem, combine its own data and institutional knowledge with frontier AI, operate within a clear regulatory framework and turn the result into a service that is commercially viable, secure and trusted? That, to me, is where the real competition begins.
The AI race in finance is therefore unlikely to produce one dominant financial centre. It is more likely to produce a network of specialised centres, each offering a different combination of capital, regulation, technology, infrastructure, talent and trust. And, that is why, coming out of GFF 2026, I would look beyond the obvious financial capitals.
I would watch Dubai because it is explicitly attempting to make itself AI-native. I would watch Malaysia because its technology and infrastructure story is becoming increasingly consequential. I would watch Labuan because its specialised international-finance role could become more valuable as cross-border financial activity becomes increasingly AI-enabled. And, I would watch New Zealand because its emphasis on adoption, application and institutional trust offers an alternative model to the scale-driven AI race.
The future of AI in finance may not belong to the jurisdiction that talks most loudly about artificial intelligence. It may belong to the one that quietly builds the regulatory, technological and institutional infrastructure that allows intelligence to become finance.
Read the LinkedIn post
By FSCL Director Riaz Patel
Why Dubai, Malaysia, Labuan and New Zealand may matter more to the next phase of AI-enabled financial services than their size suggests
There was something in the Global Fintech Fest 2026 discussion around artificial intelligence that I found more significant than the familiar debate about efficiency. The real question is no longer how AI can make financial institutions faster or cheaper, but what happens when intelligence itself becomes part of financial infrastructure.
If AI can make specialised knowledge cheaper, faster and more accessible, where will the next generation of AI-enabled financial services actually be built? My instinct is that the answer will not necessarily come only from London, New York, Singapore or Hong Kong. I would watch Dubai, Malaysia, Labuan and New Zealand precisely because each offers a different combination of technology, regulation, international connectivity, infrastructure and institutional trust.
The distinction matters because AI is increasingly becoming an ecosystem question. The financial centre of the future will depend not simply on banks and fintech companies, but on regulators, data, computing infrastructure, cybersecurity, legal frameworks, talent, capital and the ability to convert advanced technology into regulated and trusted financial activity.
Dubai is perhaps the most explicit experiment. The Dubai International Financial Centre has announced its ambition to become the world's first AI-native financial centre, with AI embedded across regulation, infrastructure and talent. Its proposed full-stack AI Campus is intended to bring together regulation, training, compute and physical AI, while DIFC also plans to provide financial firms with access to advanced AI tools.
What interests me is the underlying proposition. Dubai is not merely inviting banks to come and use AI. It is asking what a financial centre would look like if its legal, regulatory, technological and talent infrastructure were designed around AI from the beginning.
That could become a meaningful competitive advantage. The next generation of financial centres may be judged less by how many institutions they host and more by how effectively they connect institutions with technology, regulators, data and capital.
The challenge for Dubai will be execution. An AI-native financial centre cannot ultimately be judged by the ambition of its strategy. It will be judged when an AI system moves from analysing information to initiating an action, and the institution has to demonstrate who authorised that action, what information the system used, what controls applied and who remains accountable.
Malaysia offers a different proposition and its opportunity lies in connecting financial services with the rapidly expanding technology and physical infrastructure surrounding AI. Malaysia has become an important Southeast Asian data-centre location, particularly around Johor, with investment from major technology companies. The scale of that expansion is becoming significant enough to create its own infrastructure questions, including electricity demand, demonstrating that AI infrastructure is ultimately a physical economic proposition rather than something that exists only in the cloud.
For me, this creates an interesting financial possibility. AI-native finance needs computing power, data centres, connectivity, cybersecurity and payment infrastructure, but it also needs institutions capable of turning that capacity into regulated financial products and services. Malaysia therefore does not have to choose between being a technology hub and being a financial hub. Its more interesting opportunity may be to connect the two.
This is where Labuan becomes particularly relevant. Labuan does not need to compete with Dubai by attempting to become another global financial capital. Its more compelling role could be much more specialised: A cross-border financial environment in which AI can be applied to treasury, wealth structures, insurance, investment vehicles, Islamic finance and other international financial activities.
The opportunity, as I see it, is to make specialised expertise more accessible. A smaller international financial centre does not necessarily need enormous teams performing every legal, compliance, analytical, treasury and risk function if increasingly capable AI systems can help institutions access that expertise more efficiently. The competitive advantage could therefore lie in orchestration rather than scale.
Malaysia provides the broader technology and infrastructure environment. Labuan can potentially provide a specialised international-finance layer within that environment. If the two are connected intelligently, the combination could become considerably more interesting than either proposition considered in isolation.
New Zealand represents another model altogether. It is a smaller financial market, geographically distant from the major Asian centres and without the scale of India, Malaysia or the UAE. That is precisely why I find its approach worth watching. New Zealand's national AI strategy deliberately emphasises adoption and application rather than trying to compete in foundational AI development.
There is considerable realism in that approach. Not every country needs to build a frontier AI model, just as every financial centre does not need to become another Silicon Valley. There is economic value in becoming exceptionally good at applying available AI safely and intelligently to banking, insurance, payments, wealth management, compliance and financial advice.
New Zealand also introduces something that will become increasingly valuable as AI becomes more autonomous - institutional trust. An AI system operating in finance will eventually have to answer questions that go beyond whether it works. What was it permitted to access? What decision did it make? Under whose authority? Could the decision be reviewed? Who is responsible if the system is wrong?
These questions become much more consequential when AI moves from assisting people to acting on their behalf. That transition is already becoming visible. GFF 2026 itself placed agentic AI alongside tokenisation and quantum technologies as technologies capable of reshaping financial infrastructure. India is also moving towards increasingly sophisticated AI-enabled financial systems, while regulators are simultaneously placing greater emphasis on data responsibility, consumer protection and supervisory capability.
At GFF, RBI Governor Sanjay Malhotra stressed that data should be treated as a fiduciary responsibility, while SEBI has said it is moving towards adoption of a global supervisory toolkit for AI. This is important because the financial value of AI will ultimately depend on trust.
An AI system that assists a compliance officer is one thing. An AI agent capable of initiating a payment, reallocating a portfolio or making a financial recommendation is something else. Once machines begin acting with financial authority, identity, permissions, auditability and liability become part of the technology itself. That could become one of the most important competitive distinctions between financial jurisdictions.
Dubai has the ambition to build around this future. Malaysia has the physical and technological infrastructure opportunity. Labuan has a specialised cross-border financial proposition. New Zealand offers a useful model of adoption accompanied by institutional trust and responsible governance.
I would therefore resist the temptation to view these jurisdictions as competitors fighting for the same market. They could instead become complementary components of a future financial architecture.
A financial institution could potentially use Dubai for international capital and market access, Malaysia for technology and regional connectivity, Labuan for specialised cross-border structures, and New Zealand for particular governance, innovation or market applications. The future financial centre may consequently be less a single location than a network of specialised jurisdictions.
That is also where I return to the idea of headroom. AI can create headroom by reducing the time spent on routine analysis, documentation, reconciliation, coding and other knowledge-intensive activities. But headroom is not itself a competitive advantage. It becomes one only when institutions use the additional capacity to do things that were previously too expensive, too complicated or too specialised to undertake.
A financial centre that uses AI merely to reduce operating costs may become more efficient. A financial centre that uses AI to make previously uneconomic financial services possible could become structurally more competitive.
There is, of course, a significant risk in this transition. AI is creating new financial-stability questions around concentration in technology providers, enormous investment in computing infrastructure and the possibility that financial institutions become dependent on a relatively small number of external technology platforms. The more deeply AI becomes embedded in finance, the more regulators will have to understand not only how banks use AI but how financial markets themselves become exposed to the infrastructure and companies powering it.
That is why I believe jurisdictional diversity matters. Malaysia can test the intersection of manufacturing, infrastructure and finance. Labuan can explore specialised cross-border financial applications. Dubai can push the AI-native financial-centre proposition. New Zealand can demonstrate how advanced technology can be adopted without sacrificing institutional trust.
And there is a broader lesson for financial centres watching this transition from the sidelines. I would not measure success by the number of AI funds announced, AI districts established or memoranda of understanding signed. Those may be useful signals, but they are not the real test, which is much harder.
Can a financial institution take a genuinely difficult problem, combine its own data and institutional knowledge with frontier AI, operate within a clear regulatory framework and turn the result into a service that is commercially viable, secure and trusted? That, to me, is where the real competition begins.
The AI race in finance is therefore unlikely to produce one dominant financial centre. It is more likely to produce a network of specialised centres, each offering a different combination of capital, regulation, technology, infrastructure, talent and trust. And, that is why, coming out of GFF 2026, I would look beyond the obvious financial capitals.
I would watch Dubai because it is explicitly attempting to make itself AI-native. I would watch Malaysia because its technology and infrastructure story is becoming increasingly consequential. I would watch Labuan because its specialised international-finance role could become more valuable as cross-border financial activity becomes increasingly AI-enabled. And, I would watch New Zealand because its emphasis on adoption, application and institutional trust offers an alternative model to the scale-driven AI race.
The future of AI in finance may not belong to the jurisdiction that talks most loudly about artificial intelligence. It may belong to the one that quietly builds the regulatory, technological and institutional infrastructure that allows intelligence to become finance.
Read the LinkedIn post