The Algorithm Will See You Now: Malaysia Bets on AI-Powered Banking
Picture this: you photograph your electricity bill, speak a casual instruction in Manglish (“Pay this lah”), and watch as an artificial intelligence system parses the image, extracts the payment details, and completes the transaction in seconds. No app navigation. No account numbers. No authentication dance with one-time passwords.
This isn't speculative technology. It's Ryt Bank, Malaysia's first fully AI-powered financial institution, which launched to the public on 25 August 2025. Built on ILMU, the country's first homegrown large language model developed by YTL AI Labs in collaboration with Universiti Malaya, Ryt Bank represents something far more consequential than another digital banking app. It's a fundamental rethinking of the relationship between humans and their money, powered by conversational AI that understands not just English and Bahasa Melayu, but the linguistic hybrid of Manglish and even regional dialects like Kelantanese.
The stakes extend far beyond Malaysia's borders. As the world's first AI-native bank (rather than a traditional bank retrofitted with AI features), Ryt Bank is a living experiment in whether ordinary people will trust algorithms with their financial lives. The answer could reshape banking across Southeast Asia and beyond, particularly in emerging markets where digital infrastructure has leapfrogged traditional banking channels.
But here's the uncomfortable question underlying all the breathless press releases and promotional interest rates: are we witnessing genuine financial democratisation, or simply building more sophisticated systems that will ultimately concentrate power in the hands of those who control the algorithms?
The Digital Banking Gold Rush
To understand Ryt Bank's significance, you need to grasp the broader transformation sweeping through Malaysia's financial landscape. In April 2022, Bank Negara Malaysia (BNM), the country's central bank, issued five digital banking licences, deliberately setting out to disrupt a sector that had grown comfortably oligopolistic. The licensed entities included GXBank (backed by Grab), Boost Bank, AEON Bank, KAF Digital Bank, and Ryt Bank, a joint venture between YTL Digital Capital and Singapore-based Sea Limited.
The timing was strategic. Malaysia already possessed the infrastructure foundations for digital financial transformation: 97% internet penetration, 95% smartphone ownership, and 96% of adults with active deposit accounts, according to Bank Negara Malaysia data from 2024. The country had surpassed its 2026 digital payment target of 400 transactions per capita ahead of schedule, reaching 405 transactions per capita by 2024. What was missing wasn't connectivity but innovation in how financial services were delivered and experienced.
The results have been dramatic. GXBank, first to market, accumulated 2.16 billion ringgit (approximately 489 million US dollars) in customer deposits within the first nine months of 2024, becoming the largest digital bank by asset size at 2.4 billion ringgit by September 2024. Boost Bank, launching later, had attracted 399 million ringgit in assets within its first three months of operations.
Yet awareness hasn't automatically translated to adoption. Of the 93% of Malaysians who reported awareness of digital banks in Q4 2024, only 50% had actually become users. This gap reveals something crucial: people remain uncertain about entrusting their money to app-based financial institutions, particularly those without physical branches or familiar brand legacies.
Ryt Bank entered this cautious market with a differentiator: AI so deeply integrated that the bank's entire interface could theoretically be conversational. No menus to navigate. No forms to fill. Just talk to your bank like you'd talk to a financially savvy friend.
The Intelligence Behind the Interface
ILMU, the large language model powering Ryt Bank's AI assistant, represents a significant technological achievement beyond its banking application. Developed by YTL AI Labs, ILMU is designed to rival global AI leaders like GPT-4 whilst being specifically optimised for Malaysian linguistic and cultural contexts. In Malay MMLU benchmarks (which test language model understanding), ILMU reportedly outperforms GPT-4, DeepSeek V3, and GPT-5, particularly in handling regional dialects.
This localisation matters profoundly. Global AI models trained predominantly on English-language internet content often stumble when encountering the linguistic complexity of multilingual societies. Malaysia operates in at least three major languages (Bahasa Melayu, English, and Mandarin), plus numerous regional variations and the unique creole of Manglish. A banking AI that understands “I want to pindah duit to my mak's account lah” (mixing Malay, English, and colloquial structure) is genuinely useful in ways that a generic chatbot translated into Malay would never be.
The technical architecture allows Ryt AI to handle transactions through natural conversation in text or voice, process images to extract financial information (bills, receipts, payment QR codes), and provide spending insights by analysing transaction patterns. During the early access period, users reported completing full account onboarding, including electronic Know Your Customer (eKYC) verification, in approximately two minutes.
But technical sophistication creates new vulnerabilities. Every AI interaction involves sending potentially sensitive financial data to language model systems that process, interpret, and act on that information. Dr Adnan Zaylani Mohamad Zahid, Assistant Governor of Bank Negara Malaysia, has articulated these concerns explicitly. In a July 2024 speech on banking in the era of generative AI, he outlined risks including AI model bias, unstable performance in self-learning systems, third-party dependencies, data privacy vulnerabilities, and emerging cyber threats like AI-enabled phishing and deepfakes. His message was clear: “Human judgment must remain central to risk management oversight.”
The Trust Equation
Trust in financial institutions is a peculiar thing. It's simultaneously deeply rational (based on regulatory frameworks, deposit insurance, historical performance) and thoroughly emotional (shaped by brand familiarity, peer behaviour, and gut instinct). AI banking disrupts both dimensions.
On the rational side, Ryt Bank is licensed by Bank Negara Malaysia and protected by Perbadanan Insurans Deposit Malaysia (PIDM), which guarantees deposits up to 250,000 ringgit per depositor. Yet according to 2024 global banking surveys, 58% of banking customers across 39 countries worry about data security and hacking risks. Only 28% believe their bank effectively communicates data protection measures, and only 40% fully trust their bank's transparency about cybersecurity.
These trust deficits are amplified when AI enters the picture. Research on consumer trust in AI financial services reveals that despite technological sophistication, adoption “hinges significantly on human trust and confidence.” Malaysia isn't immune to these anxieties. A TikTok user named sherryatig captured the sentiment bluntly when commenting on Ryt Bank: “The current banking system is already susceptible to fraud. NOT in my wildest dream to allow transactions from prompt.”
The regional context intensifies these worries. Consumers across Southeast Asia hold banks and fintech firms primarily responsible for safeguarding against financial crimes, and surveys indicate that more than half of respondents across five Southeast Asian markets expressed growing fears about rising online fraud and hacking.
Yet early Ryt Bank user reviews suggest cautious optimism. Coach Alex Tan praised the “smooth user experience” and two-minute onboarding. Tech reviewers noted that “even in beta, Ryt AI is impressively intuitive, making banking feel less like a task and more like a conversation.” The AI's ability to process screenshots of bank account details shared via WhatsApp and automatically populate transfer fields has been highlighted as solving a genuine pain point.
These positive early signals, however, come from early adopters who tend to be more tech-savvy and risk-tolerant than the broader population. The real test will come when Ryt Bank attempts to expand beyond enthusiastic technophiles to the mass market, including older users, rural communities, and those with limited digital literacy.
The Personalisation Paradox
One of AI banking's most touted benefits is hyper-personalisation: financial services tailored precisely to individual circumstances, goals, and behaviour patterns. The global predictive analytics market in banking is forecast to grow at a compound annual growth rate of 19.42% through 2030. Bank of America's Erica virtual assistant, which uses predictive analytics, has over 19 million users and reportedly generated a 28% increase in product adoption compared to traditional marketing approaches.
This sounds wonderful until you examine the underlying dynamics. Personalisation requires extensive data collection and analysis. Every transaction, every app interaction, every moment of hesitation before clicking “confirm” becomes data that feeds the AI's understanding of you. The more personalised your banking experience, the more comprehensively you're surveilled.
Moreover, AI-driven personalisation in financial services has repeatedly demonstrated troubling patterns of bias and discrimination. An analysis of Home Mortgage Disclosure Act data from the Urban Institute in 2024 revealed that Black and Brown borrowers were more than twice as likely to be denied loans compared to white borrowers. Research on fintech algorithms found that whilst they discriminated 40% less than face-to-face lenders, Latinx and African-American groups still paid 5.3 basis points more for purchase mortgages and 2.0 basis points more for refinance mortgages compared to white counterparts.
These disparities emerge because AI models learn from historical data that encodes past discrimination. The technical challenge is compounded by what researchers call the “fairness paradox”: you cannot directly measure bias against protected categories without collecting data about those categories, yet collecting such data raises legitimate concerns about potential misuse.
Bank Negara Malaysia has acknowledged these challenges. The central bank's Chief Risk Officers' Forum developed an AI Governance Framework outlining responsible AI principles, including fairness, accountability, transparency, and reliability. In August 2025, BNM unveiled its AI financial regulation framework at MyFintech Week 2025 and initiated a ten-week public consultation period (running until 17 October 2025) seeking feedback on sector-specific AI definitions, regulatory clarity needs, and AI trends that could shape the sector over the next three to five years.
But regulatory frameworks often lag behind technological deployment. By the time comprehensive AI banking regulations are finalised and implemented, millions of Malaysians may already be using systems whose algorithmic decision-making remains opaque even to regulators.
The Inclusion Question
Digital banks, including AI-powered ones, have positioned themselves as champions of financial inclusion, promising to serve the underserved. The rhetoric is appealing, but does it match reality?
Malaysia's financial inclusion challenges are substantial. According to the 2023 RinggitPlus Malaysian Financial Literacy Survey, 71% of respondents could save 500 ringgit or less monthly, whilst 67% had emergency savings lasting three months or less. The Khazanah Research Institute reports that 55% of Malaysians spend equal to or more than their earnings, living paycheck to paycheck. Approximately 15% of the 23 million Malaysian adults remain unbanked, according to The Business Times. MSMEs face a particularly acute 90 billion ringgit funding gap.
Bank Negara Malaysia data indicates that close to 60% of customers at GXBank, AEON Bank, and Boost Bank come from traditionally underserved segments, including low-income households and rural communities. Boost Bank's surveys in Kuala Terengganu found that 97% of respondents did not have 2,000 ringgit readily available.
However, digital banks face inherent limitations in reaching the truly marginalised. One of the primary challenges is bridging the digital divide, particularly in underserved communities where many individuals and businesses, especially in rural areas, lack necessary devices and digital literacy. Immigrants and refugees often lack the documentation required for digital identity verification. Elderly populations may struggle with smartphone interfaces regardless of how “intuitive” they're designed to be.
There's also an economic tension in AI banking's inclusion promise. Building and maintaining sophisticated AI systems requires substantial ongoing investment. Those costs must eventually be recovered through fees, product cross-selling, or data monetisation. The business model that supports free or low-cost AI banking may ultimately depend on collecting and leveraging user data in ways that create new forms of exploitation, even as they expand access.
Ryt Bank launched with 4% annual interest on savings (on the first 20,000 ringgit, until 30 November 2025), unlimited 1.2% cashback on overseas transactions with no conversion fees, and a PayLater feature providing instant credit up to 1,499 ringgit with 0% interest if repaid within the first month. These are genuinely attractive terms. But as reviews have noted, “long-term value will depend on whether these benefits are extended after November 2025.” The pattern is familiar from countless fintech launches: aggressive promotional terms to build user base, followed by monetisation pivots.
The Human Cost of Efficiency
AI banking promises remarkable efficiency gains. Chatbots and virtual assistants can handle up to 50% of customer inquiries, according to industry estimates. Denmark's DNB bank reported that within six months, its chatbot had automated over 50% of all incoming chat traffic and interacted with over one million customers.
But efficiency has casualties. Across Southeast Asia, approximately 11,000 bank branches are expected to close by 2030, representing roughly 18% of current physical banking presence. In Malaysia specifically, strategy consulting firm Roland Berger projects nearly 567 bank branch closures by 2030, a 23% decline from 2,467 branches in 2020 to approximately 1,900 branches.
These closures disproportionately affect communities that already face financial service gaps. Rural areas lose physical banking presence. Elderly customers who prefer face-to-face service, immigrants who need in-person assistance, and small business owners who require relationship banking all find themselves pushed toward digital channels they may neither trust nor feel competent to use.
The employment implications extend beyond branch closures. By the end of 2024, 71% of banking institutions and development financial institutions had implemented at least one AI application, up 56% from the previous year. Each of those AI applications represents tasks previously performed by humans. Customer service representatives, loan officers, fraud analysts, and financial advisers increasingly find their roles either eliminated or transformed into oversight positions managing AI systems.
Industry estimates suggest AI could generate between 200 billion and 340 billion US dollars annually for banking. Yet there's a troubling asymmetry: those efficiency gains and cost savings accrue primarily to financial institutions and shareholders, whilst job losses and service degradation are borne by workers and vulnerable customer segments.
The Algorithmic Black Box
Perhaps the most profound challenge AI banking introduces is opacity. Traditional banking, for all its faults, operates on rules that can theoretically be understood, questioned, and challenged. AI systems, particularly large language models like ILMU, operate fundamentally differently. They make decisions based on pattern recognition across vast training datasets, identifying correlations that may not correspond to any human-comprehensible logic. Even the engineers who build these systems often cannot fully explain why an AI reached a particular conclusion, a problem known in the field as the “black box” dilemma.
This opacity has serious implications for financial fairness. If an AI denies you credit, declines a transaction, or flags your account for fraud investigation, can you meaningfully challenge that decision? Consumer complaints about banking chatbots reveal experiences of “feeling stuck and frustrated, receiving inaccurate information, and paying more in junk fees” when systems malfunction or misunderstand user intent.
Explainability is considered a core tenet of fair lending systems, yet may work against AI adoption. America's legal and regulatory structure to protect against discrimination and enforce fair lending “is not well equipped to handle AI,” according to legal analyses. The Consumer Financial Protection Bureau has outlined that financial institutions are expected to hold themselves accountable for protecting consumers against algorithmic bias and discrimination, but how regulators can effectively audit systems they don't fully understand remains an open question.
Bank Negara Malaysia's approach has been to apply technology-agnostic regulatory frameworks. Rather than targeting AI specifically, existing policies like Risk Management in IT (RMiT) and Management of Customer Information and Permitted Disclosures (MCIPD) address associated risks comprehensively. The BNM Regulatory Sandbox facilitates testing of innovative AI use cases, allowing supervised experimentation.
Yet regulatory sandboxes, by definition, exist outside normal rules. The question is whether lessons learned in sandboxes translate to effective regulation of AI systems operating at population scale.
The Cyber Dimension
AI banking's expanded attack surface introduces new cybersecurity challenges. According to research on AI cybersecurity in banking, 80% of organisational leaders express concerns about data privacy and security, whilst only 10% feel prepared to meet regulatory requirements. The areas of greatest concern for financial organisations are adaptive cyberattacks (93% of respondents), AI-powered botnets (92%), and polymorphic malware (83%).
These aren't theoretical threats. Malware specifically targeting mobile banking apps has emerged across Southeast Asia. ToxicPanda and TgToxic, which emerged in mid-2022, target Android mobile users with bank and finance apps in Indonesia, Taiwan, and Thailand. These threats will inevitably evolve to target AI banking interfaces, potentially exploiting the conversational nature of systems like Ryt AI to conduct sophisticated social engineering attacks.
Consider the scenario: a user receives a message that appears to be from Ryt Bank's AI assistant, using familiar conversational style and regional dialect, requesting confirmation of a transaction. The user, accustomed to interacting with their bank via natural language, might not scrutinise the interaction as carefully as they would a traditional suspicious email. AI-enabled phishing could exploit the very user-friendliness that makes AI banking appealing.
Poor data quality poses another challenge, with 40% of respondents citing it as a reason AI initiatives fail, followed by privacy concerns (38%) and limited data access (36%). An AI banking system is only as reliable as its training data and ongoing inputs. Corrupted data, whether through malicious attack or simple error, could lead to widespread incorrect decisions.
What Happens When the Algorithm Fails?
Every technological system eventually fails. Servers crash. Software has bugs. Networks go offline. In traditional banking, these failures are inconvenient but manageable. But what happens when an AI-native bank experiences a critical failure?
If ILMU's language processing system misunderstands a transaction instruction and sends your rent money to the wrong account, what recourse do you have? If a software update introduces bugs that cause the AI to provide incorrect financial advice, who bears responsibility for decisions made based on that advice?
These questions aren't adequately addressed in current regulatory frameworks. Consumer complaints about banking chatbots show that whilst they're useful for basic inquiries, “their effectiveness wanes as problems become more complex.” Users report “wasted time, feeling stuck and frustrated” when chatbots cannot resolve issues and no clear path to human assistance exists.
Ryt Bank's complete dependence on AI amplifies these concerns. Traditional banks and even other digital banks maintain human customer service channels as fallbacks. If Ryt Bank's differentiator is comprehensive AI integration, building parallel human systems undermines that efficiency model. Yet without adequate human backup, users become entirely dependent on algorithmic systems that may not be equipped to handle edge cases, emergencies, or their own malfunctions.
The phrase “computer says no” has become cultural shorthand for the frustrating experience of being denied something by an inflexible automated system with no human override. AI banking risks creating “algorithm says no” scenarios where financial access is controlled by systems that cannot be reasoned with, appealed to, or overridden even when obviously wrong.
The Sovereignty Dimension
An underappreciated aspect of ILMU's significance is technological sovereignty. For decades, Southeast Asian nations have depended on Western or Chinese technology companies for critical digital infrastructure. Malaysia's development of a homegrown large language model capable of competing with global leaders like GPT-4 represents a strategic assertion of technological independence.
This matters because AI systems encode the values, priorities, and cultural assumptions of their creators. A language model trained predominantly on Western internet content will inevitably reflect Western cultural norms. ILMU's deliberate optimisation for Bahasa Melayu, Manglish, and regional dialects ensures that Malaysian linguistic and cultural contexts are centred rather than accommodated as afterthoughts.
The geopolitical implications extend further. As AI becomes infrastructure for financial services, healthcare, governance, and other critical sectors, nations that control AI development gain significant strategic advantages. Malaysia's ILMU project demonstrates regional ambition to participate in AI development rather than remaining passive consumers of foreign technology.
However, technological sovereignty has costs. Maintaining and advancing ILMU requires sustained investment in AI research, computing infrastructure, and talent development. Malaysia must compete globally for AI expertise whilst building domestic capacity.
Ryt Bank's use of ILMU creates a testbed for Malaysian AI at scale. If ILMU performs reliably in the demanding environment of real-time financial transactions involving millions of users, it validates Malaysia's AI capabilities and could attract international attention and investment. If ILMU encounters significant problems, it could damage credibility and confidence in Malaysian AI development more broadly.
The Question of Control
Ultimately, the transformation AI banking represents is about control: who controls financial data, who controls access to financial services, and who controls the algorithms that increasingly mediate between people and their money.
Traditional banking, for all its inequities and exclusions, distributed control across multiple points. Bank employees exercised discretion in lending decisions. Regulators audited and enforced rules. Customers could negotiate, complain, and exert pressure through collective action. The system was far from perfectly democratic, but power wasn't entirely concentrated.
AI banking centralises control in the hands of those who design, train, and operate the algorithms. Those entities (corporations, in Ryt Bank's case the YTL Group and Sea Limited partnership) gain unprecedented insight into user behaviour, financial circumstances, and potentially even personal lives, given how much can be inferred from transaction patterns. They decide what features to build, what data to collect, which users to serve, and how to monetise the platform.
Regulatory oversight provides some counterbalance, but regulators face profound information asymmetries. They lack the technical expertise, computational resources, and internal access necessary to fully understand or audit complex AI systems. Even when regulators identify problems, enforcement mechanisms designed for traditional banking may be inadequate for addressing algorithmic harms that manifest subtly across millions of automated decisions.
The power imbalance between individual users and AI banking platforms is even more stark. Terms of service that few users read grant broad rights to collect, analyse, and use personal data. Algorithmic decision-making operates opaquely, with limited user visibility into why particular decisions are made. When problems occur, users face AI systems that may not understand complaints and human support channels that are deliberately limited to reduce costs.
Financial exclusion can cascade into broader life exclusion: difficulty renting housing, accessing credit for emergencies, or even proving identity in an increasingly digital society. If AI systems make errors or biased decisions, the affected individuals often have limited recourse.
The Path Forward
So will Malaysia's first AI-powered bank fundamentally change how ordinary people manage their money and trust financial institutions? The answer is almost certainly yes, but the nature of that change remains contested and uncertain.
In the optimistic scenario, AI banking delivers on its promises. Financial services become more accessible, affordable, and personalised. Underserved communities gain banking access that traditional institutions never provided. AI systems prove trustworthy and secure, whilst regulatory frameworks evolve to effectively address algorithmic risks. Malaysia demonstrates that developing nations can be AI innovators rather than passive technology consumers.
This scenario isn't impossible. The technological foundations exist. Regulatory attention is focused. Public awareness of both benefits and risks is growing. If stakeholders act responsibly and prioritise long-term sustainability over short-term gains, AI banking could genuinely improve financial inclusion and service quality.
But the pessimistic scenario is equally plausible. AI banking amplifies existing inequalities and creates new forms of exclusion. Algorithmic bias reproduces and scales historical discrimination. Data privacy violations and security breaches erode trust. Job losses and branch closures harm vulnerable populations. The concentration of power in AI platforms creates new forms of corporate control over economic life. The promised benefits accrue primarily to young, urban, digitally literate users whilst others are left behind.
This scenario isn't dystopian speculation. It reflects documented patterns from fintech and platform economy deployments worldwide. The optimistic and pessimistic scenarios will likely coexist, with AI banking simultaneously creating winners and losers.
What's most important is recognising that technological change isn't inevitable or predetermined. The impact of AI banking will be shaped by choices: regulatory choices about what to permit and require, corporate choices about what to build and how to operate it, and individual choices about what to adopt and how to use it.
Those choices require informed public discourse that moves beyond both techno-optimism and techno-pessimism to engage seriously with the complexities and trade-offs involved. Malaysians shouldn't simply accept AI banking as progress or reject it as threat, but rather interrogate it critically: Who benefits? Who is harmed? What alternatives exist? What safeguards are necessary?
The Conversation We Need
Ryt Bank's conversational AI interface is designed to make banking feel natural, like talking to a financially savvy friend. But perhaps what Malaysia most needs isn't a conversation with an algorithm, but a conversation amongst citizens, regulators, technologists, and financial institutions about what kind of financial system serves the public interest.
That conversation must address uncomfortable questions. How much privacy should people sacrifice for convenience? How much human judgment should be replaced by algorithmic efficiency? How do we ensure that AI systems serve the underserved rather than just serving themselves? Who bears responsibility when algorithms fail or discriminate?
The launch of Malaysia's first AI-powered bank is genuinely significant, not because it provides definitive answers to these questions, but because it makes them urgently tangible. Ryt Bank is no longer speculation about AI's potential impact on banking but a real system that real people will use to manage real money and real lives.
Early user reviews suggest that the technology works, that the interface is intuitive, that transactions happen smoothly. But technology working isn't the same as technology serving human flourishing. The question isn't whether AI can power a bank (clearly it can) but whether AI banking serves the public good or primarily serves corporate and technological interests.
Bank Negara Malaysia's public consultation on AI in financial services, running until 17 October 2025, represents an opportunity for Malaysians to shape regulatory approaches whilst they're still forming. But effective participation requires moving beyond the promotional narratives of frictionless, intelligent banking to examine the power structures and social implications underneath.
The 93% of Malaysians who are aware of digital banks but remain cautious about adoption aren't simply being backward or technophobic. They're exercising appropriate scepticism about entrusting their financial lives to systems they don't fully understand, controlled by entities whose interests may not align with their own.
That scepticism is valuable. It should inform regulatory design that insists on transparency, accountability, and human override mechanisms. It should shape corporate strategies that prioritise user control and data privacy over maximum data extraction. It should drive ongoing research into algorithmic bias, security vulnerabilities, and unintended consequences.
AI banking will change how Malaysians manage money and relate to financial institutions. But whether that change is fundamentally positive or negative, inclusive or exclusionary, empowering or exploitative remains to be determined. The algorithm will indeed see you now, but the crucial question is: are you being seen clearly, fairly, and on terms that serve your interests rather than merely its own?
The answer lies not in the technology itself but in the social, political, and ethical choices that surround its deployment. Malaysia's experiment with AI-powered banking is just beginning. How it unfolds will offer lessons far beyond the country's borders about whether artificial intelligence in finance can genuinely serve human needs or ultimately subordinates those needs to algorithmic logic.
That's the conversation worth having, and it's one that no AI, however sophisticated, can have for us.
Sources and References
Bank Negara Malaysia. (2022). “Five successful applicants for the digital bank licences.” Retrieved from https://www.bnm.gov.my/-/digital-bank-5-licences
Bank Negara Malaysia. (2020). “Policy Document on Licensing Framework for Digital Banks.” Retrieved from https://www.bnm.gov.my/-/policy-document-on-licensing-framework-for-digital-banks
Zahid, Adnan Zaylani Mohamad. (2024, July 16). “Banking in the era of generative AI.” Speech by Assistant Governor of Bank Negara Malaysia. Bank for International Settlements. Retrieved from https://www.bis.org/review/r240716g.htm
TechWire Asia. (2025, January). “Malaysia's first AI-powered bank revolutionises financial services.” Retrieved from https://techwireasia.com/2025/01/malaysia-first-ai-powered-bank-revolutionises-financial-services/
SoyaCincau. (2025, August 12). “Ryt Bank First Look: Malaysia's first AI-powered Digital Bank.” Retrieved from https://soyacincau.com/2025/08/12/ryt-bank-ytl-digital-bank-first-look/
Fintech News Malaysia. (2025). “Ryt Bank Debuts as Malaysia's First AI-Powered Digital Bank.” Retrieved from https://fintechnews.my/53734/digital-banking-news-malaysia/ryt-bank-launch/
YTL AI Labs. (2025). “YTL Power Launches ILMU, Malaysia's First Homegrown Large Language Model.” Retrieved from https://www.ytlailabs.com/
New Straits Times. (2025, August). “YTL launches ILMU – Malaysia's first multimodal AI, rivalling GPT-4.” Retrieved from https://www.nst.com.my/business/corporate/2025/08/1259122/ytl-launches-ilmu-malaysias-first-multimodal-ai-rivalling-gpt-4
TechNode Global. (2025, March 21). “RAM: GXBank tops Malaysia's digital banking customer deposits with $489M for first nine months of 2024.” Retrieved from https://technode.global/2025/03/21/ram-gxbank-tops-malaysias-digital-banking-customer-deposits-with-489m-for-first-nine-months-of-2024/
The Edge Malaysia. (2024). “GXBank tops digital banking sector deposits with RM2.16 bil as of September 2024 – RAM Ratings.” Retrieved from https://theedgemalaysia.com/node/748777
The Edge Malaysia. (2024). “Banking for the underserved.” Retrieved from https://theedgemalaysia.com/node/727342
RinggitPlus. (2023). “RinggitPlus Malaysian Financial Literacy Survey 2023.”
Roland Berger. (2020). “Banking branch closure forecast for Southeast Asia.”
Urban Institute. (2024). “Home Mortgage Disclosure Act data analysis.”
MX. (2024). “Consumers Trust in AI Integration in Financial Services Is Shifting.” Retrieved from https://www.mx.com/blog/shifting-trust-in-ai/
Brookings Institution. “Reducing bias in AI-based financial services.” Retrieved from https://www.brookings.org/articles/reducing-bias-in-ai-based-financial-services/
ResearchGate. (2024). “AI-Powered Personalization In Digital Banking: A Review Of Customer Behavior Analytics And Engagement.” Retrieved from https://www.researchgate.net/publication/391810532
Consumer Financial Protection Bureau. “Chatbots in consumer finance.” Retrieved from https://www.consumerfinance.gov/data-research/research-reports/chatbots-in-consumer-finance/
Cyber Magazine. “How AI Adoption is Challenging Security in Banking.” Retrieved from https://cybermagazine.com/articles/how-ai-adoption-is-challenging-security-in-banking
No Money Lah. (2025, August 27). “Ryt Bank Review: When AI meets banking for everyday Malaysians.” Retrieved from https://nomoneylah.com/2025/08/27/ryt-bank-review/
Tim Green UK-based Systems Theorist & Independent Technology Writer
Tim explores the intersections of artificial intelligence, decentralised cognition, and posthuman ethics. His work, published at smarterarticles.co.uk, challenges dominant narratives of technological progress while proposing interdisciplinary frameworks for collective intelligence and digital stewardship.
His writing has been featured on Ground News and shared by independent researchers across both academic and technological communities.
ORCID: 0009-0002-0156-9795 Email: tim@smarterarticles.co.uk