Money Frontier 2026 Countdown 10 Days: Summit Highlights Preview
Money Frontier 2026 (https://www.moneyfrontier.info/zh) will be held at the Hopewell Centre in Hong Kong from July 27 to 28, bringing together industry leaders, top platforms, investors, policy researchers, and frontline operators to share real observations and practical experience from the markets, products, and industry.
The summit will not merely discuss the macro trends of Web3 and AI; it will focus on the concrete changes that are already happening and reshaping the industry, helping attendees gain clarity on the new market structure, understand where opportunities arise and risks lie hidden, and formulate executable judgments.
As the public markets and crypto markets accelerate their convergence, why is the price of capital becoming more inconsistent? As traditional high-return opportunities gradually disappear, where can investors still seek opportunities with manageable risk? As leading AI companies have become difficult to invest in or are overvalued, have ordinary investors truly missed this wave?
Agenda Highlights at a Glance
Highlight One: Latest Regulatory Policy Developments
U.S. Strategic Bitcoin Reserve, What's Next?
The Bitcoin Policy Institute, which has long been involved in U.S. Bitcoin policy research and advocacy, will share the latest progress on the U.S. Strategic Bitcoin Reserve and related policies. This organization consistently conducts Bitcoin research and public policy initiatives targeting policymakers and continuously publishes research related to strategic reserves.
What is worth noting is not just whether the Bitcoin Act will pass, but the specific institutional path the strategic reserve will take, the source of the assets, how Congress and the executive branch will coordinate, and how these policy changes might affect institutional allocation, market liquidity, and digital asset policies in other countries. This session will help attendees look beyond the headlines to understand the actual pace of policy advancement, key obstacles, and potential market impacts.
Members of the Hong Kong Legislative Council will also share their perspectives on the future development direction of digital assets in Hong Kong.
Highlight Two: Understanding the New Market Structure
As DeFi and CeFi continue to develop, Digital Asset Treasury companies (DATs), RWAs, asset tokenization, and new asset issuance platforms are accelerating the connection between public markets and crypto markets.
But this connection has not unified the markets; instead, it has amplified liquidity fragmentation. Due to differences between platforms in capital costs, access conditions, collateral rules, redemption mechanisms, trading hours, and jurisdictions, the same asset can have different interest rates, prices, and liquidity across different platforms and between digital asset and traditional financial markets.
The Bank for International Settlements has pointed out that RWA and tokenization could exacerbate market fragmentation and increase financing costs. Research from the Federal Reserve also shows that cross-market pricing frictions and market segmentation persist between digital currency markets and traditional financial markets.
Therefore, capital allocation is no longer just about choosing a platform and comparing headline yields. It requires identifying frictions and boundaries between different markets, understanding different product structures, and discerning whether price discrepancies stem from market access, liquidity, credit, leverage, subsidies, or poorly identified risks. Consequently, managing capital costs, redemption, custody, counterparty, smart contract, and regulatory risks becomes paramount.
On July 27, founders and CEOs of leading protocols including Ethena, Spark (MakerDAO), and others will deeply analyze their product structures, yield sources, and platform operating mechanisms. The CEO and Chairman of Strive will also discuss new Bitcoin-based credit products like STRC and SATA, sharing their structures, yield logic, and potential risks.
Several senior executives from traditional financial institutions and the digital asset industry will also provide frontline observations and unique insights from different market and business perspectives.
Highlight Three: When AI Makes Everyone a Target Worth Attacking
As AI moves from concept to reality, security issues are no longer limited to large institutions, trading platforms, or high-net-worth individuals.
Identities, accounts, assets, communication records, and social connections can all become entry points for attacks. Faced with cheaper, more scalable, and highly personalized attack methods, are traditional security habits still effective?
The summit will discuss how AI is changing the way attackers select targets, gather information, and execute attacks, and how individuals and institutions should re-evaluate identity verification, asset custody, device permissions, and internal security processes.
The core question is not just "Is AI dangerous?", but: When the cost of attack drops rapidly, how should we increase the cost for attackers?
Highlight Four: Finding Opportunities and Growth Paths in the AI Wave
Worried you've already missed the AI wave? Not necessarily. The better entry point for AI investment might not be before a technological breakthrough, but when the technology is validated by the market and rapid growth begins to expose industrial bottlenecks.
For most investors, opportunities may not come from betting early on the next big AI application. Investing significant time and capital to judge an unproven technology often means assuming risks where one holds no advantage.
A more realistic path is to wait for the technology or trend to complete market validation and enter a phase of rapid expansion, then look for industrial bottlenecks exposed during this growth. As demand surges, key resources like GPUs, memory, data centers, and electricity often cannot scale synchronously, leading to supply-demand imbalances.
Compared to predicting who the next winner will be, these links already validated by real demand but constrained by supply capacity may offer clearer judgment criteria and participation paths for ordinary investors.
Which shortages are just temporary cyclical mismatches? Which bottlenecks might persist for years? Which assets, while part of the AI industry chain, cannot truly share in its growth? How can one identify key links with pricing power, barriers to expansion, and real customer demand?
You will find answers to these questions in the July 27 agenda.
Notable investors like Starmap and others will share their investment experience and judgment frameworks in the frontier technology primary market; Xiandao will also continue the "Second Life Practical Manual," sharing how to translate judgments on new trends into executable personal choices and practices.
Day Two: From GPU to Electricity, Deconstructing the Full AI Computing Infrastructure Chain
Highlight One: Domestic Chips Accelerate Entry into Intelligent Computing Centers, Opening New Industry Windows
With the accelerated rollout of large model inference and industry AI applications, the competition for domestic computing power is no longer limited to a single technological path. How can domestic general-purpose GPUs further penetrate intelligent computing centers and real business scenarios? How can China leverage its industrial chain advantages to overtake and shake the market position of overseas GPU giants? Can ASICs optimized for specific AI tasks achieve a breakthrough balancing performance, energy efficiency, cost, and large-scale deployment? The summit will invite representatives from Moore Threads and Dr. Yang Zuoxing, founder of YANJI Electronics, a company deeply engaged in domestic AI ASIC R&D that has launched the self-developed "Shenmu" brand, to share R&D progress, practical implementation, and industrialization directions of domestic intelligent computing chips from different technological paths, collectively observing the new opportunities opening up in the local computing ecosystem.
Highlight Two: Domestic Open-Source Large Models Lower the Barrier to Innovation, Agents Move Toward Real-World Applications
With the advent of high-performance open-source large models like ChatGLM 5.2 and Kimi K3, more companies can directly access model capabilities close to the frontier. Models are no longer a proprietary resource for a few leading companies. The competitive focus of the AI industry is shifting from "who can train a larger model" to "who can build truly usable products and systems based on the model."
This also opens a new development window for AI Agents. When model capabilities become a callable fundamental resource, how can enterprises further connect data, tools, and business processes? Which Agents have moved beyond concept demonstrations into real production scenarios? Which applications have sustainable demand, willingness to pay, and potential for large-scale replication? How can model capabilities, computing costs, and business models form a closed loop?
Through topics such as "The Agent Wave: Comprehensive Reshaping from Technological Evolution to a New Industry Era" and "From Model to Agent: The Implementation, Computing, and Capitalization Path of AI-Native Applications," the summit will invite guests from Tencent Cloud, BytePlus Hong Kong, KUAI.CLOUD, as well as AI startups and investment institutions, to share real-world cases and development directions of Agents. They will observe the new generation of AI application ecosystems fostered by domestic open-source models from different dimensions, including technology evolution, product implementation, computing power support, and commercialization paths.
Highlight Three: From Data Center Globalization to AI Factory, Engineering Capability Becomes the Core Barrier
AI data centers are not simply about adding GPUs to traditional server rooms. As power density per rack continues to increase, power supply and distribution, cooling, network interconnection, equipment deployment, and operation systems all need to be redesigned. The ability to organize land, electricity, equipment, and operational capabilities into a stable, efficient, and sustainably scalable system is becoming the genuine engineering barrier of the computing power industry.
Data center globalization also brings more complex real-world problems: How do you select sites and power conditions suitable for AI workloads? How do you control construction timelines and delivery costs? What differences exist between markets in infrastructure, supply chains, and operating environments? What upgrades are needed for traditional data centers to truly evolve into "AI Factories" geared for AI training and inference?
Around topics like "Data Center Globalization" and "Beyond Traditional Data Centers: The Rise of AI Factories and Intelligent Computing Power," companies including Canaan, Xinke Intelligent, Skyward Digital, JDK Capital, and Goodvision AI, as well as industry guests like the head of a special task force under Korea's Presidential AI National Strategy Commission, will deconstruct the key aspects of data center planning, construction, globalization, and operation based on frontline project experience. They will discuss which experiences are replicable, which technologies and construction risks are most easily overlooked. The agenda will also focus on core links such as power supply reliability, rack power density, cooling systems, network interconnection, and operational capabilities.
Highlight Four: From POW to AI, the Value Boundaries of Power Resources Are Being Redefined
If GPUs determine the performance of computing power, and data centers determine how that computing power is hosted, then electricity constitutes the most fundamental resource constraint of the entire computing system. As AI computing power demand continues to grow, AI data centers and POW are competing for the same scarce resources: stable and cost-advantageous electricity, suitable sites for deploying high-density equipment, and power contracts that can lock in long-term costs.
In this context, electricity is no longer just an operational cost for a data center. It could become a strategic asset that needs to be independently allocated, operated, and re-valued. Can different types of computing loads be flexibly switched based on market demand? Can existing POW infrastructure further support AI computing? How can the computing value generated per megawatt of power resources be improved? What new asset forms and business models might emerge around electricity, sites, and load scheduling?
Through topics like "From POW to AI" and "What Problems Did POW Solve? What is the Next Problem?", the summit will extend from existing computing power industry experience to the allocation efficiency, yield flexibility, and future opportunities of power resources. Whether existing AI computing services or Token distribution mechanisms can achieve efficient, transparent, and scalable resource organization, similar to how mining pools allocate hashing power, remains to be further validated by the industry. Starting from the problems POW has solved, KuPool will also explore the new propositions the computing power industry needs to face: when chips and models constantly iterate, what truly defines the boundaries of industry expansion might not just be the power resources themselves, but also the ability to organize, schedule, and trade computing power.
For more details, please visit: https://www.moneyfrontier.info/
For participation and cooperation inquiries, please contact the summit staff.
