Futures
Access hundreds of perpetual contracts
CFD
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Pre-IPOs
Unlock full access to global stock IPOs
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
Promotions
AI
Gate AI
Your all-in-one conversational AI partner
Gate AI Bot
Use Gate AI directly in your social App
GateClaw
Gate Blue Lobster, ready to go
Gate for AI Agent
AI infrastructure, Gate MCP, Skills, and CLI
Gate Skills Hub
10K+ Skills
From office tasks to trading, the all-in-one skill hub makes AI even more useful.
GateRouter
Smartly choose from 40+ AI models, with 0% extra fees
Taipei International Computer Show » Jensen Huang arrives in Taiwan on 5/27, securing a spot as TSMC's Southern Science Park AI chip production capacity becomes the focus
Nvidia CEO Jensen Huang is expected to arrive in Taiwan on May 27 and, on May 28, will hold the “Trillion-Dollar Banquet,” widely inviting major leaders from Taiwan’s supply chain. The market is focusing on whether Nvidia will lock in TSMC’s advanced packaging capacity in the Southern Taiwan Science Park.
Nvidia CEO Jensen Huang is about to make a whirlwind visit to Taiwan. Reports say he is scheduled to arrive in Taiwan on May 27 and hold the “Trillion-Dollar Banquet” on May 28, inviting heavyweight players from Taiwan’s supply chain. The market is watching whether Nvidia will further secure TSMC’s advanced packaging capacity in the Southern Taiwan Science Park, plan for mass production of next-generation AI GPUs after 2028, and team up with MediaTek to showcase new edge device chips.
Jensen Huang’s “Trillion-Dollar Banquet” guest list revealed! Taiwan’s supply chain gears up for Computex
The Commercial Times has revealed Huang’s Taiwan itinerary. He will host the “Trillion-Dollar Banquet” on the 28th, inviting Taiwan supply-chain leaders such as TSMC, Foxconn, Delta Electronics, MediaTek, and others to a dinner, warming up for this annual tech event.
Huang is expected to deliver a keynote speech on June 1 ahead of the official opening of COMPUTEX 2026. He will focus on the next stage of AI development, covering computing, physical AI, and AI agent systems, and will fully explain the AI infrastructure blueprint from energy, networks, chips, and systems to applications.
Nvidia already positioned itself with TSMC for advanced packaging; the Southern Taiwan Science Park becomes a key hub for AI chip production
In addition, the market is paying close attention to whether Huang during this trip will meet directly with TSMC executives to finalize details of long-term cooperation on advanced process capacity. In fact, Nvidia’s strategy of tightly aligning with TSMC has already been put into action.
Earlier, Chain News reported that in TSMC’s Chiayi AP7 advanced packaging plant, Plant P2 focuses on SoIC packaging, with Nvidia as its main customer, reaching a monthly capacity of 12,000 units; the Southern Taiwan Science Park’s AP8 plant is an important base for CoWoS packaging, with expected monthly capacity to exceed 40,000 units by the end of 2026. CoWoS and SoIC technologies are the key advanced packaging processes supporting Nvidia’s continued improvement in GPU computing performance.
On the wafer manufacturing side, Area A of TSMC’s Southern Taiwan Science Park development zone will be planned as the Fab 22 P7 plant. Construction is expected to start in the second quarter. The plant will target 2nm and more advanced processes and reserve expansion space for two additional fabs. The plant will take on preparations for the mass production of Nvidia’s next-generation “Feynman” series chips, together with the existing 3nm and 5nm processes at Fab 18 and the AP8 advanced packaging plant. With this, the Southern Taiwan Science Park has formed a complete AI chip production cluster, becoming the core base for Nvidia’s next-generation GPU supply chain.
When Huang visited Taiwan last November, his first stop was Tainan. It was also reported that he had expressed interest in investing in land reserved near TSMC’s Fab 18—specifically the P10 and P11 reserved sites. All these moves clearly reveal Nvidia’s strategic intent to deploy ahead of schedule for TSMC’s Southern Taiwan capacity.
Looking at the tech giants’ wave of “de-TSMC-ization,” it further highlights Nvidia’s strategic value in staying deeply tied to TSMC
While Nvidia deepens its cooperation with TSMC, other major tech companies are also starting a trend of spreading risk. Tesla has already signed a long-term AI chip supply contract with Samsung extending to 2033; AMD CEO Lisa Su has personally gone to Samsung’s Pyeongtaek plant in South Korea to deepen collaboration; Apple has also reportedly been cultivating Intel (Intel) as a backup foundry responsible for contract manufacturing across its entire product line.
This wave of “de-TSMC-ization” does not appear to shake Nvidia’s existing approach. Instead, it further underscores the strategic significance of Nvidia choosing to continue its deep binding with TSMC. In the AI era, requirements for yield and supply stability are far higher than before. Any process misstep could lead to large-scale shipment delays, and TSMC’s technological leadership in advanced processes and advanced packaging remains an advantage that is still difficult to replace.
Industry analysts believe that as AI GPU computing demand continues to surge, advanced processes and advanced packaging have gradually shifted from the past “supply-chain coordination” model to a “capacity reservation” and “long-term binding” model. In particular, because the build cycles for processes below 2nm and advanced packaging such as CoWoS and SoIC are longer, major global AI chip makers have already reserved capacity for the next 3 to 5 years to avoid supply-demand shortages.
Stepping into Taiwan’s supply chain, Nvidia’s ecosystem integrates deeply
As demand for AI servers, GB300, Vera Rubin, liquid cooling, power supplies, network switches, and advanced packaging expands across the board, Taiwan’s supply chain role within Nvidia’s ecosystem continues to be upgraded. TSMC holds key advanced process and packaging capacities such as CoWoS and SoIC; Foxconn, Quanta, and Wistron focus on complete AI server systems and rack-level solutions; Delta Electronics benefits from the trend toward high-power power supplies, cooling, and 800V DC architectures; MediaTek is also expected to team up with Nvidia at COMPUTEX 2026 to jointly showcase new-generation edge device chips.
Jensen Huang’s visit is not only to warm up for the annual tech event, but also an important observation indicator for the global AI capital expenditure cycle. The market is also highly focused on Nvidia’s latest financial report that it is set to release, including the shipment progress of Blackwell, data center revenue performance, and the mass production timeline for the next-generation Rubin platform.