On May 19, 2026, the "Smarter AI, Safer Digital Wealth" summit, hosted by the Asia Academy of Digital Economics (AADE) and co-organized by ONERHT, the Responsible Fintech Institute, and the Blockchain Association of Singapore (BAS), was grandly held at the Shaw Foundation Alumni House of the National University of Singapore.
The summit brought together ambassadors and diplomatic envoys to Singapore from eight countries, including Australia, Spain, India, Ireland, and Panama. Guests included representatives from INTERPOL, the Monetary Authority of Singapore (MAS), as well as top international and Asian financial, crypto asset, and legal institutions such as Visa, Standard Chartered Bank, HashKey, StraitsX, SBI DM, QCP, and RHTLaw Asia, engaging in a multi-dimensional dialogue on digital wealth security.
View the full data recap: https://aades.academy/deai26.html
Opening | Singapore's Three Cards: Technology, Hub, and Trust
Delivering the opening address was Dr. Neo Kok Beng, Nominated Member of Parliament (NMP) of Singapore and Chief Scientist of the Asia Academy of Digital Economics. He summarized Singapore's positioning in the AI era into three points: striving for global leadership in emerging technologies (AI, quantum computing), advocating that small nations must dare to "place bets" and treat failure as experimentation; serving as a global AI hub, not by competing on the largest models and computing power, but on integration, application, and innovation; and continuing to act as a global connector.
Among the three, he elaborated most on the third, underpinned by one word: trust. In his view, the connector's value lies not in the technology itself, but in "the trust that you will deliver, honor agreements, abide by the law, and uphold the principles and values you have signed onto." Evidence of this is the recently signed critical supplies mutual supply agreement between New Zealand and Singapore: Singapore committed to not imposing export restrictions on diesel and fuel oil, despite not producing oil itself, a confidence rooted in its hub status and delivery reliability. He summarized this capability as "resilience."
He also casually pointed out another thread for the day: payments today are invisible; even billionaires hold nothing more than a string of numbers in a banking app. When wealth is completely digitized and placed in custody, the basis for judging whether it is "safe" shifts from visible assets to invisible trust.
This was a proposition the subsequent agenda returned to repeatedly. After assets converge into on-chain digits and decision-making and execution are delegated to AI and autonomous agents, "trust" is no longer diplomatic rhetoric but an engineering problem that must be answered separately by governance frameworks, by code and consensus, and by continuous monitoring.
01 | Safer? On-Chain Governance in the AI Era
Roundtable | The TradFi Dilemma: Are We Still Building Web3, or Rebuilding Traditional Finance with Blockchain?
Moderated by Thomas Wan, General Manager of the Blockchain Association of Singapore (BAS), the discussion first set the tone using the Gartner Hype Cycle: Web3 has passed through the peak of inflated expectations and the trough of disillusionment (the FTX collapse and the Mango Markets exploit were tuition fees), and now, driven by the US GENIUS Act and CLARITY Act, is entering a more pragmatic phase. However, the convergence of AI + Web3 + digital assets could lead to systemic risks if "agentic AI" is taken lightly. The core question was just one: With KYC, whitelisting, and centralized interventions entering the scene, are we building Web3, or just draping traditional finance in a blockchain veneer?
Aaron Chua, a member of the BAS Regulatory Sub-Committee, answered by saying "don't create a binary opposition": the regulatory principle is always "risk-based." Web3 has evolved from decentralized protocols/DEXs to centralized exchanges, and now to TradFi beginning to embed smart contract infrastructure designs. The direction of this spectrum is to bring the "good things" of Web3 (decentralization, smart contracts, programmable logic, embedded compliance) into Web2 and TradFi, and then build new products on top. It's not about "forgetting Web3 and going back to Web2," but about bringing Web2's governance into Web3 to create something that combines the strengths of both.
Daniel Yang, Head of Compliance at QCP, viewed institutionalization as a "natural evolution": Early Web3 in 2010 indeed brought accessibility, low cost, and better settlement, but it also bred bad actors, though the scale was negligible compared to today's capital flows. When even nation-state actors start using Web3 for attacks and arbitrage, coupled with the FTX scandal in the 2020s and the collapse of a tokenized exchange in Singapore last year, the public naturally asks, "Is my money afforded the same level of security?" Therefore, as institutions enter, "some degree of control" is inevitable; but the opportunity lies in "using technology itself to install the right controls, reducing friction while returning the security standards, costs, and accessibility of the traditional world back to the end-user."
Amanda Chen, Partner at RHTLaw Asia, qualified to practice in both China and Singapore, offered the clearest judgment criterion of the entire session. Her core point was: Web3.0 must defend self-custody and permissionless innovation, and the key to judging whether a new regulation is "protecting the market" or "protecting incumbents" lies in whether the rule applies equally to all participants.
Specifically, if a rule requires "governance must be transparent and auditable," this sounds strict, but why is it still "protective" rather than a "power guard"? Because whether it's a large bank or a small startup, everyone must meet the same standard. Although the cost is high, the rule itself is the same for everyone. Similarly, "stablecoin issuers must meet X capital requirements" — if new issuers and existing banks use the same standard, it is equal; "all users can use decentralized dispute resolution" — the mechanism is open to everyone.
But if a rule says "only regulated and certified banks can run nodes," that is different. On the surface, it sounds like a security requirement, but in reality, it excludes certain people's eligibility to participate at the rule level. No matter how hard a small startup tries or how much money it spends, it is almost impossible to meet the condition of "becoming a bank." The design of the rule itself presupposes who can play and who cannot. Similarly, "regulators give large institutions real-time data interfaces, but nothing to small players," or "large institutions can freeze transactions via backdoor APIs, while small users have no such power." These are rules that appear neutral on the surface but are actually tailored for vested interests.
Her judgment criterion boils down to: Does this rule, technically and institutionally, open the same opportunities to everyone? If the answer is "Yes, although difficult, everyone can try," then it is a truly protective rule; if the answer is "No, the design of this rule presupposes that only certain people can do it," then it becomes a new protective cloak for power. The name has changed to code, but the essence remains unchanged. So, every time a new regulation is introduced, she suggests asking a simple question: Is this rule setting a high threshold that everyone can cross, or is it building a gate for some that only they can enter?
Sanchit Mall, Director of Crypto & Digital Currency Business for Asia Pacific at Visa, offered a more "pipeline" perspective: For Visa, the core is payment and fund flows. By attaching a card to a crypto asset/stablecoin, it can be spent anywhere Visa is accepted, backed by years of accumulated trust. Rather than getting tangled in the divide between TradFi and Web3, it's better to think about how to piece the two together.
Faced with the moderator's sharpest question — Web3 was originally meant to end the old order where "a few giants call the shots," but now the market is dominated by a handful of players. Isn't this just centralized finance in a new guise, betraying Web3's original intent?
Daniel Yang did not directly defend "decentralization" but reframed this concern as "growing pains." His logic was: Today's landscape is not a "regression to the old ways" but rather "not yet grown up." TradFi has banks, clearing houses, custodians, and payment networks; Web3 actually has corresponding roles — custodians, stablecoin liquidity providers, on-chain analytics firms — just under different names. The reason there are so few players is that, relative to the vast TradFi market, this market is still too small and in its early stages.
It is worth noting that he essentially shifted the question from "Have we betrayed the ideal?" (a value judgment) to "Has this market matured?" (a stage judgment). And to let it continue growing, he pointed to three things: first, talent, which is severely lacking and requires people who truly understand blockchain; second, PPP (public-private partnerships), where enterprises, vendors, BAS, and regulators must sit at the same table; third, the speed of compliance must catch up with the speed of business and innovation. The good news is that the world is flooding in: traditional giants like Visa and various banks have recently entered the field, and crypto-native players are also moving towards the traditional end. In his view, the more players there are, the healthier the competition, and the ultimate beneficiaries will not just be the financial elites well-served by TradFi, but every ordinary person.
The moderator's next question harbored a more common expectation: Could permissioned DeFi be the very bridge that allows institutions to smoothly enter DeFi? As a "bridge builder" straddling TradFi and DeFi, Ong Chun Kiat (CK), Acting CEO of SBI Digital Markets, answered very bluntly: No.
What he wanted to debunk was the common misunderstanding of the "bridge." Most people think building a bridge means "packaging an institutional-grade product and putting it on-chain" (i.e., tokenization). CK said this is precisely what won't work: you are selling a financial asset, not a consumer product; there are legal obligations, a complete asset lifecycle, and regulations that must be followed behind it. Tokenization alone cannot solve these problems. The real difficulty lies not in "moving TradFi things on-chain," but within DeFi itself. DeFi must actively evolve, elevating platform and public chain standards to "institutional grade." On-chain natives must also accept what they generally resist today: KYC-style account opening and access control. In other words, whether the bridge can be built depends on whether DeFi is willing to converge towards TradFi's standards, not the other way around, and "we are still far from that step."
However, he left a gentle closing note: Web3 has never been a lawless wilderness; its essence is using technology to serve assets, allowing payments and assets to flow freely across accounts and borders. Even if the complete bridge hasn't been built yet, every asset and every function moved on-chain is already a tangible step forward for both Web2 and Web3.
The Final Question: What Must Be Defended?
In the final round, the host asked everyone to name "the one thing that must be preserved to achieve institutional trust and scaled adoption without losing the decentralized soul":
lAmanda Chen: Unilateral exitThe ability to leave one system for another with your self-custodied assets, without anyone's permission. Web3 isn't just about being fast and cheap; it's about eliminating "gatekeepers who can arbitrarily deny you service." A permissioned system can still be considered Web3 as long as exit is possible; but once exit requires a committee vote, a signature from a licensed custodian, or compliance with a blacklist you can't see or appeal, it is no longer Web3—just "traditional finance with a hash."
lAaron Chua: Transaction traceability and transparencyWithout it, peer-to-peer transfers and self-custody lose their meaning, and integrity is impossible.
lSanchit Mall: InteroperabilityThere are many chains, stablecoins, crypto assets, and payment agents; they must be able to flow between one another. Visa aims to be that bridge, preserving Web3's uniqueness while ensuring it is well-governed.
lOng Chun Kiat: Programmable trustPut everything on-chain so that asset holding and transfer do not rely on any counterparty; regulators, issuers, and investors must all trust the same token, and everything you program on that token must be trustworthy.
lDaniel Yang: BalancePreserve Web3's promise of innovation and experimentation while installing appropriate controls; the biggest taboo is treating Web3 digital assets and traditional financial assets with a "one-size-fits-all" approach, which would stifle genuinely beneficial innovation. This field is still very new, and regulation is still evolving. Rather than a knee-jerk pile-on of controls, it should be proportional and risk-based.
Keynote | Jason Tay: Web3's Autonomous AI Arms Race—Building Not Stronger Robots, but a "Biological Immune System"
HashKey OTC Singapore CEO Jason Tayspoke about two sides of the same coin: autonomous agents are like fire—they can cook your food, or burn down your mother-in-law's house.
He first presented a "bad news as good news" statistic: According to blockchain security firm Hacken, Web3 losses due to exploits in Q1 2026 amounted to approximately $464 million. While staggering, this was the lowest quarterly loss since 2023 (phishing and smart contract vulnerabilities accounted for the bulk). Compared to the $1.46 billion Bybit hack in 2025, $464 million is just a fraction. More noteworthy is the trend: losses are shifting from "single catastrophic events" to "mid-sized incidents spread across multiple protocols." Why? Because today's AI defense bots are better at "invisible defense."
Among the "invisible intermediaries," there are good actors. For example, automated smart contract auditing: Manual auditing is like finding an invisible needle in a burning haystack, taking weeks for complex DeFi protocols; autonomous agents using symbolic execution can complete the run in minutes. AI-enhanced auditing tools can now identify up to 85% of common vulnerabilities (such as integer overflows) before a single gwei is spent.
There are also bad actors. The same technology used to find bugs to fix can also find bugs to exploit. Zero-day vulnerabilities (where the vendor is completely unprepared, with exactly zero days left to patch), fuzzing (bombarding contracts with random data to find openings), and Ethereum's "dark forest": MEV bots lurking to front-run profitable transactions (he gave an analogy: you're queuing to buy the last croissant, and a Terminator teleports in front of you, buys it, and sells it back to you at double the price, all before you can say "I'll be back").
Thus, this sub-second digital arms race: "Our bots versus their bots." But the deadlock of the arms race lies here: The defender must be 100% correct every single time; the attacker only needs to be right once.So "building better bots" is not the answer. What needs to change is the architecture of the battlefield itself: shifting from reactive responses to possessing one's own "biological immune system," like a self-healing network, completing detection, pausing, and countermeasures before an exploit is finished.
He concluded: Code is neutral; speed is not.The moment a smart contract is finalized on-chain, autonomous agents have already read the code, simulated the outcomes, and mapped the vulnerabilities, while the human operator hasn't even refreshed the dashboard. We cannot win a 21st-century architecture war using human supervision. The two milestones for Web3 security are building open, collaborative, self-healing protocols—a "shared immune system" serving DeFi. The age of machines has arrived; our job is to be the "Guardians of the Galaxy."
02 | Smarter? But How to Build Trust in the Age of Intelligence...
Keynote | David Hardoon: From "Framework Scarcity" to "Framework Surplus," AI Governance Needs a Unified Language
Former Chief Data Officer of the Monetary Authority of Singapore (MAS) and former Global Head of AI Enablement at Standard Chartered Bank, David Hardoon, highlighted an irony overlooked by the industry.
Back in 2017 and 2018, when he was at MAS, there were virtually no AI governance guidelines on the market. The FEAT principles (Fairness, Ethics, Accountability, Transparency) they spearheaded were initially not even for regulation, but for "development": the industry said, "If you don't give us guidelines, we dare not do anything," so the regulator first provided guardrails.
Seven or eight years later, the pendulum has swung to the other extreme. Hardoon cited OECD data: the number of tracked national-level AI governance frameworks has now surpassed 900, spanning roughly 18 jurisdictions, not including the hundreds of playbooks issued by various standards bodies, consulting firms, and industry players. The problem that followed: clarity has not increased; instead, fragmentation, overlap, conflicting standards, and decision fatigue have emerged. The question for boards is no longer "how to mitigate risk," but "which framework should we even use?" In finance and Web3, where autonomous systems are already deployed and operating at machine speed, this confusion is no longer an academic issue.
He also pointed out a "double standard" phenomenon. Every time AI is to be deployed, the CISO (Chief Information Security Officer) and compliance officers throw out a long list of "buts": cannot connect to SharePoint (Note: SharePoint is Microsoft's enterprise collaboration and document management platform, often containing a company's vast internal and even sensitive data. So when compliance says "don't let AI access SharePoint," the subtext is the fear that AI, once connected, might access or leak these internal files), security must meet standards, this and that must be satisfied. Hardoon said he agrees with these requirements 100%, but he often asks in return: "And how well are we humans doing on that front today?" The response is usually silence. The implication: when the same security risks occur with humans, institutions have long "seen and tacitly accepted" them, continuing operations as usual; yet when it comes to AI, they demand it meets a standard that even humans cannot reach.
Ultimately, this is a question of how to view operational risk. Either the yardstick is the same for humans and AI, or we must admit we are applying a double standard.
The solution he offered is not complicated, borrowing an old concept from engineering cybernetics. Governance can be divided into three layers: Why (the ethical foundation), What (regulatory obligations, risk classification, cross-border considerations), How (increasingly, operational manuals from vendors and standards bodies). But one piece is missing above these three layers: a socio-technical foundation that can measure them and unify "people, rules, and AI." His core proposition is: governance cannot have "one set for AI, one for people, one for rules," but must be unified governance.
What does unified governance measure? The same enduring quantities from engineering: Observability (can we see what the agent is thinking and doing), Controllability (can a human or regulator take over at any time, press that big red button), Stability (will it start oscillating under pressure), Robustness (can it withstand adversarial prompts or market shocks), and Performance (is it actually delivering). These were previously discussed in their own silos (cybersecurity, adversarial attacks, performance, supervision). What he aims to do is bring them together, remaining neutral on "human or AI." He noted that some might question: can you "observe" a human the way you monitor a machine? Hardoon said we have actually been doing this for a long time, just under different names—KPIs, quarterly reviews. The real concern is another scenario: if we layer defenses only on the AI side while neglecting to monitor the "human" side, risk will specifically exploit this weak link. He used an analogy: just as a hacker who can't breach the building's front door might sneak in through the unguarded boiler room system, AI governance is the same—once the "human" gap is left open, risk can still find its way in.
His final point drove the theme home: governance should shift from being "a checkpoint you must pass by trekking through the desert every 60 days" to continuous monitoring, because what is safe today may not be safe tomorrow. Simplified, implementable, unified, and no more double standards.
Keynote | Zhu Feida: When Knowledge Becomes Infrastructure, the Best Intelligence is Not "Artificial Intelligence"
Singapore Management University (SMU) School of Computing Associate Professor and Vice Dean Zhu Feidachose a "less noticed" angle: turning the most valuable yet hardest-to-solidify asset in a company—experience—into reusable "intelligence assets."
He first described a scenario: a colleague who resigned two weeks ago suddenly pops up in your message box: "I'm back, we can still work together as before, except now I'm an agent, part of the company's digital infrastructure." This is not science fiction: in some companies, 30%–40% of employeesalready exist in the form of digital twins/infrastructure. This raises a series of thorny questions: If over 30% of work is done by AI, how is employee value assessed? How are privacy, consent, and dignity safeguarded (especially valued in Singapore)? Does knowledge contributed to a process belong to the company, the employee, or the AI? Job design needs to be deconstructed: clearly defining which part AI handles, which part the human handles, and how the two seamlessly connect.
He proposes a "new data stack": the cloud provides scalable storage, data becomes the foundation for AI consumption, AI provides scalable intelligence, platforms stitch them together, and Web3 introduces trust and provenance, making everything "programmably trustworthy." The result is that enterprises shift from "process-driven" to "intelligence-driven": in the past, competition was about process consistency, discipline, and SOPs, but in the future, the more competitive companies will be those whose knowledge is more "learnable."
Here he proposes "second-order thinking": when encountering a new problem, don't rush to ask "how do I solve this."
Take a step back, deconstruct the entire process of "how I approach this problem" (what to do first, where to find data, how to piece it together), structure it into a skill, and hand it to AI. Next time, AI will do it faster for you.
When the tacit know-how that lives in our brains and in informal firefighting discussions is turned into shareable, portable, versionable, plug-and-play modules like Lego blocks, it becomes an "AI-ready" asset.
His most interesting point is this: the best intelligence is never just artificial intelligence, but a combination of artificial intelligence + human intelligence + organizational intelligence.
He first used reCAPTCHA as a clever example: this verification system was originally designed by professors at Carnegie Mellon, allowing users to identify words while logging in—seemingly just to verify you are human. But behind the scenes, millions of document images were quietly annotated this way, and human visual cognition was thus absorbed into the computing system. This is the first layer of fusion: the human brain did what machines couldn't do well, and the machine preserved this capability for the human brain.
Today's more complex AI systems follow the same logic: they don't just rely on data itself, but also integrate human judgment and perception. For example, in news aggregation, combining the discussion heat on X, the frequency of celebrity coverage, and reader comments and rankings, what is finally presented to you is not just "facts," but a kind of "narrative intelligence," incorporating the active choices of human aesthetics, experience, and values.
But this is not enough. When this "intelligence + human perception" becomes a core asset, organizational intelligence is needed—that is, a system of trust and incentives. Who owns this data? Who can use it? Who benefits? How to distribute fairly? These questions become incredibly real. And this is precisely the problem Web3 aims to solve: if AI gives us scalable intelligence, Web3 gives us a scalable trust mechanism. The future competition is not just about whose system is smarter, but whose system is more trustworthy.
Roundtable | Flipping the Question: What Can Finance Do for AI?
This roundtable was moderated by Dr James Ong, founder of the Artificial Intelligence International Institute (AIII). He flipped the usual question: everyone asks what AI and Web3 can do for finance, but rarely asks "what can finance do for AI and Web3."
StraitsX Co-founder and CEO Tianwei Liu started with stablecoins: what truly drives stablecoin adoption is the global (especially emerging market) rigid demand for USD exposure beyond local currencies; when account opening and receiving payments are nearly zero-cost, "Bitcoin neobanks" and card products linked to Visa/Mastercard make inclusive finance commercially viable for the first time. His take on "agentic payment" is very down-to-earth: people talk a lot about machine-to-machine, API-to-API, but what truly improves life is "buy me a bubble tea, buy coffee for the whole team."
Vest Capital Managing Partner Riady Gozali emphasized the creation, coordination, and distribution of value: AI scales intelligence, Web3 provides trust and coordination, and finance is responsible for "amplifying responsibly"; the next wave of value will come more from intangible assets like data, human expertise, and relationship networks.
The sharpest insights came from Professor David Lee Kuo Chuen, Chairman of APEX (Asia Pacific Exchange) and Chairman of the Global Fintech Institute. He reminded us that this is the first time in human history that we are delegating "decision-making and execution power" to "another species," and that "letting AI govern AI" is unfeasible because AI can lie and hide; the only viable path is to use Web3 and the philosophy of Bitcoin to immutably record everything AI has done, so that AI cannot move a step when conditions are not met. He even threw out a counter-intuitive viewpoint: we must ensure that AI (agents) themselves are inefficient, just as Bitcoin processes only single-digit transactions per second but uses massive computing power to resist malicious attacks and build trust; by making the virtual space sufficiently "inefficient," humans can be efficient enough using AI, otherwise humanity will be in trouble.
03 Back to the Human | "Alien Digital Immigrants" and the Anxiety of Every Profession
The keynote speech by Dr Pei Sai Fan, former Banking Department Head at the Monetary Authority of Singapore (MAS), was the most philosophical. Borrowing American technology ethicist Tristan Harris's metaphor, he called advanced AI "alien digital immigrants." He posed questions on four levels: Humanity (the erosion of human agency and autonomy, the weakening of social trust, but also the opportunity to liberate people from production and redefine meaning), Social Institutions (our entire institutional framework is built on the assumption of "scarcity," which is now invalid, and "abundance" may be more destructive than scarcity), Market Mechanisms (markets excel at allocating resources under scarcity and rewarding efficiency, but are not designed for dignity and social cohesion; KPIs may need to expand from "productivity contribution" to "social contribution"), and Education (Harvard, UChicago, Stanford, NTU are all interdisciplinarily weaving humanities, ethics, and AI literacy into their curricula; education is not a vocational training institute, but the gatekeeper of human civilization).
The Chinese-language session (Roundtable 3), co-moderated by Dr Chua Chee Lay and He Liqin, discussed "AI's challenge to every profession": Dr Chan Kwong Tung from the National Institute of Education (NIE/NTU) talked about how AI grading essays can liberate teachers from low-level cognitive labor, allowing them to focus on higher-order thinking; Dr Zhou Tongquan, founder of TENWIT Consultants, candidly admitted that AI can write in minutes the three-dimensional equations that used to take him half a day, but the lifelong liability-bearing signature of a professional engineer, after everything is "calculated entirely by AI," makes him hesitate on whether to sign; Shi Yunding, Strategic Operations Officer of Moutai House SG, Moutai's global flagship cultural experience project, worried that the overnight popularization of mature models would bring employment restructuring, but believed that after virtual satisfaction, people will miss offline, real, face-to-face consumption scenarios even more; Wang Jiancheng, Chairman of Ronghua Group, categorized the challenge into "awakening ability" and "reconstruction ability," and used "one formula for a thousand people" (data-driven, dynamic unity of commonality and individuality) to describe AI's transformation of retail and smart healthcare. Chua Chee Lay's final reflection was quite sharp: all the problems AI might cause—deception, hallucinations, etc.—humans have long had; AI has not yet started a war, and the fault for "car accidents" mostly lies not with the car but with the human. In other words, AI itself is innocent; the more advanced the AI era, the higher the demand for human quality.
Conclusion
Over the course of the day, governance, arbitrage, knowledge, and compliance each spoke their own language, but one main thread ran throughout: As technology becomes smarter, can we make society safer and more humane? From Hardoon's "unified governance," Jason Tay's "biological immune system," to Zhu Feida's "trustworthy systems," they all point to the same thing: in an era where we delegate decision-making power to "another species," the true moat is not stronger computing power, but a trust architecture that is observable, controllable, exitable, and treats humans and machines equally. This is precisely what the conference theme "Smarter AI, Safer Digital Wealth" aims to reveal.
Chen Baihui, President of the Asia Academy of Digital Economics, stated that DEAI 2026 is a platform that brings topics such as artificial intelligence, digital assets, financial security, educational transformation, and social governance into a single framework for discussion. He emphasized that while the digital economy pursues speed and efficiency, it must also value security, responsibility, and human value. As digital assets gradually enter the mainstream financial system and artificial intelligence begins to participate in decision-making and risk assessment, society needs to establish more mature governance mechanisms to ensure that technology truly serves enterprises, the public, and the broader social community.
Welcome to follow the Asia Academy of Digital Economics (AADE) LinkedIn account or subscribe to our newsletter for more high-quality industry conferences:
l Official Website: https://aades.academy/
l LinkedIn: https://www.linkedin.com/company/asia-academy-of-digital-economics
l Email: contact@aades.academy
l For newsletter subscription or other collaboration inquiries, please fill out: https://aades.academy/aade-nav-contact.html
This article is translated and organized based on the live English verbatim transcript of the DEAI 2026 International Summit (May 19, 2026, Singapore), with some content edited and condensed; guest views do not represent the stance of this media outlet.
