Money Frontier 2026 Summit to Explore AI, Crypto Market Structure, and Regulatory Trends

Money Frontier 2026 will be held at the Hopewell Hotel in Hong Kong on July 27–28, bringing together leading industry companies, top platforms, investors, policy researchers, and frontline operators to share real-world insights and practical experience from the market, products, and industry frontlines.

The summit does not focus on macro trends of Web3 and AI, but rather on concrete changes that have already occurred and are reshaping the industry, helping attendees understand the new market structure, identify how opportunities are forming, where risks lie, and develop actionable insights.

As the public market and crypto market accelerate their convergence, why are capital prices becoming even more fragmented? As traditional high-return opportunities gradually disappear, where can investors find risk-controlled opportunities? With AI leaders now difficult to access or overvalued, have ordinary investors truly missed out on this round of opportunity?

Preview the Agenda Highlights

What to watch next for the U.S. national strategic Bitcoin reserve?

The Bitcoin Policy Institute, which has long been involved in U.S. Bitcoin policy research and advocacy, will share the latest developments on a strategic U.S. Bitcoin reserve and related policies. The organization has consistently conducted Bitcoin research and public policy advocacy targeted at policymakers, and continues to publish research on strategic reserves.

What matters is not just whether the Bitcoin bill will pass, but how the strategic reserve will be implemented through institutional pathways, where the assets will come from, how Congress and the executive branch will coordinate, and how these policy changes may affect institutional allocations, market liquidity, and other countries’ digital asset policies. This session will help participants look beyond the headlines to understand the real pace of policy advancement, key obstacles, and their potential market impacts.

Hong Kong Legislative Council members will also share insights on the future direction of digital assets in Hong Kong.

As DeFi and CeFi continue to evolve, digital asset treasury companies (DAT), RWA, asset tokenization, and new asset issuance platforms are accelerating the connection between public markets and crypto markets.

But rather than unifying the market, connectivity has amplified liquidity fragmentation. Due to differences across platforms in funding costs, access requirements, collateral rules, redemption mechanisms, trading hours, and jurisdiction, the same asset may exhibit different interest rates, prices, and liquidity on various platforms, as well as between digital assets and traditional financial markets.

The Bank for International Settlements notes that RWA and tokenization may exacerbate market fragmentation and increase financing costs; research from the Federal Reserve also shows that there remain pricing frictions and market segmentation between the digital currency market and traditional financial markets.

Therefore, asset allocation is no longer just about choosing platforms and comparing surface-level returns, but about identifying friction and boundaries between different markets, understanding the structures of various products, and determining whether spreads arise from market access, liquidity, credit, leverage, subsidies, or unmarked risks—and managing funding costs, redemptions, custody, counterparty, smart contract, and regulatory risks accordingly.

On July 27, founders and CEOs of leading protocols including Ethena and Spark (MakerDAO) will provide an in-depth breakdown of their product structures, revenue sources, and platform operating mechanisms. The CEO and Chairman of Strive will also discuss the product structure, revenue logic, and potential risks of new credit products based on Bitcoin, such as STRC and SATA.

Senior executives from traditional financial institutions and the digital assets industry will also provide frontline insights and unique perspectives from different markets and business angles.

As AI moves from concept to reality, security concerns are no longer limited to large institutions, exchanges, or high-net-worth individuals.

Identity, accounts, assets, communication records, and social relationships can all become entry points for attacks. In the face of increasingly low-cost, 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 carry out attacks, as well as how individuals and institutions should reassess authentication, asset custody, device permissions, and internal security processes.

The core issue is not just "Is AI dangerous?" but: How should we increase the cost to attackers as the cost of attacks rapidly declines?

Worried you’ve already missed the AI wave? Not necessarily. The better entry point for investing in AI may not come before technological breakthroughs, but rather when the technology has been validated by the market and rapid growth begins to expose industry bottlenecks.

For most investors, the opportunity doesn’t come from predicting the next AI application. Spending significant time and capital to evaluate an unproven technology often means taking on risks for which you don’t have an advantage.

A more realistic approach is to wait for technology or trends to undergo market validation and enter a phase of rapid expansion, then identify industry bottlenecks that emerge during this growth. When demand surges, critical resources such as GPUs, memory, data centers, and electricity often fail to scale in tandem, leading to supply-demand imbalances.

Compared to predicting who will be the next winner, these segments—already validated by real demand but constrained by supply capacity—may offer retail investors clearer criteria for evaluation and more accessible pathways to participation.

Which shortages are merely temporary cyclical mismatches? Which bottlenecks may persist for years? Which assets, though part of the AI supply chain, fail to truly benefit from industry growth? And how can one identify the critical segments with pricing power, production expansion barriers, and genuine customer demand?

You’ll find the answers to these in the agenda on July 27.

Starfield and other renowned investors will share their investment experiences and frameworks for evaluating frontier technology in the primary market; Xiantao will also continue presenting "The Practical Guide to Second Life," sharing how to translate insights into emerging trends into actionable personal decisions and practices.

Highlight 1: Domestic chips are accelerating their entry into intelligent computing centers, opening a new industry window.

As large model inference and industry-specific AI applications accelerate deployment, competition in domestic computing power is no longer confined to a single technological pathway. How can domestic general-purpose GPUs further penetrate intelligent computing centers and real-world business scenarios? How can China’s industrial chain advantages be leveraged to overtake the market dominance of overseas GPU giants? Can ASICs optimized for specific AI tasks achieve new breakthroughs in performance, energy efficiency, cost, and scalable deployment? The summit will invite industry representatives from Moore Threads and Dr. Yang Zuoxing, founder of YANJI Electronics, a long-term pioneer in domestic AI ASIC R&D and the creator of the proprietary “Shenmu” brand, to share insights from diverse technological perspectives on the progress, practical applications, and industrialization directions of domestic intelligent computing chips, collectively exploring the new opportunities emerging in China’s local computing ecosystem.

Highlight 2: Domestic open-source large models lower the barrier to innovation, bringing agents into real-world applications

With the emergence of high-performance open-source large models such as ChatGLM 5.2 and Kimi K3, more and more companies can now directly access model capabilities approaching the cutting edge. Models are no longer exclusive resources for a few leading companies; the focus of competition in the AI industry is shifting from "who can train larger models" to "who can build truly usable products and systems based on these models."

This also opens a new development window for agents. As model capabilities become foundational, callable abilities, how can enterprises further connect data, tools, and business processes? Which agents have moved beyond conceptual demonstrations into real production environments? Which applications have the potential for sustained demand, payment capacity, and scalable replication? How can a closed loop be formed between model capabilities, computing costs, and business models?

The summit will bring together guests from Tencent Cloud, BytePlus Hong Kong, KUAI.CLOUD, and AI startups and investment firms to share real-world case studies and future directions of Agents, through topics such as “The Agent Wave: Comprehensive Reshaping from Technological Evolution to a New Industrial Era” and “From Models to Agents: Pathways for Deployment, Computing Power, and Monetization of AI-Native Applications,” examining how domestic open-source models are catalyzing a new generation of AI application ecosystems across dimensions including technological evolution, product implementation, computing infrastructure, and commercialization pathways.

Highlight 3: From data center expansion overseas to AI Factory, engineering capability has become the core barrier

An AI data center is not simply about adding GPUs to a traditional server room. As power density per cabinet continues to rise, power distribution, cooling, network interconnection, equipment deployment, and operational systems all need to be redesigned. The ability to integrate land, electricity, equipment, and operational capabilities into a stable, efficient, and sustainably scalable system is becoming the true engineering barrier in the computing power industry.

The overseas expansion of data centers also brings more complex practical challenges: How to select suitable campuses and power conditions for AI workloads? How to control construction timelines and delivery costs? What are the differences across markets in terms of infrastructure, supply chains, and operational environments? What upgrades are required for traditional data centers to truly evolve into “AI Factories” designed for AI training and inference?

Around topics such as "data center overseas expansion" and "Beyond Traditional Data Centers: The Rise of AI Factories and Intelligent Computing Power," companies including Canaan, Xinkexin Intelligence, Skyward Digital, JDK Capital, and Goodvision AI, along with industry guests such as the head of the special task force from the Presidential Advisory Committee on Artificial Intelligence National Strategy in South Korea, will draw on firsthand project experience to break down the key insights into planning, building, expanding overseas, and operating data centers, discussing which practices can be replicated and which technical and construction risks are most often overlooked. The agenda will also place special emphasis on core aspects such as power supply reliability, rack power density, cooling systems, network interconnectivity, and operational capabilities.

Highlight 4: The value boundary of power resources is being redefined from POW to AI

If GPUs determine the performance of computing power, and data centers determine how computing power is hosted, then electricity forms the most fundamental resource constraint for the entire computing power system. As demand for AI computing power continues to grow, AI data centers and POW are competing for the same scarce resources—stable and cost-effective electricity, suitable locations for deploying high-density equipment, and long-term power contracts that lock in costs.

Under this context, electricity is no longer merely an operational cost for data centers but may also become a strategic asset requiring independent configuration, operation, and revaluation. Can different types of computing workloads be flexibly switched based on market demand? Can existing POW infrastructure be further leveraged for AI computing? How can the computing value generated per megawatt of power resources be enhanced? What new asset forms and business models might emerge around electricity, campus infrastructure, and workload scheduling?

The summit will explore topics such as “From POW to AI” and “What problems did POW solve? What’s the next challenge?” building on existing experience in the compute power industry to extend the discussion to power resource allocation efficiency, revenue elasticity, and future opportunities. Whether existing AI computing services or token distribution mechanisms can achieve efficient, transparent, and scalable resource organization—similar to how mining pools allocate compute power—remains to be validated by the industry. KuPool will begin by examining what POW has already solved, and then delve into the new challenges the compute power industry must confront in the next phase: as chips and models continue to evolve, what truly determines the industry’s expansion boundaries may not only be power resources themselves, but also the ability to organize, schedule, and trade compute power.

For more details, visit: https://www.moneyfrontier.info/

For conference attendance and partnership inquiries, please contact the summit staff.