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
OpenAI's $18 billion chip manufacturing hits a snag: Broadcom's financing condition requires Microsoft to underwrite 40%, with mass production delayed until 2027.
According to Ju Insights’ Beating Monitoring, last fall, OpenAI and Broadcom announced their in-house AI chip with great fanfare, seeking to break free from reliance on NVIDIA. But this deal is now stuck on the question of “who will pay.” According to an internal memo obtained by The Information, the two sides are locked in a tug-of-war over the production funding of $18 billion for the first phase. Broadcom’s key upfront condition is that Microsoft must step in to commit to buying roughly 40% of the chips.
The chipmaking plan, codenamed Nexus, is extremely ambitious, with the ultimate goal of deploying chips to support 10 GW of power consumption (with production costs alone exceeding $180 billion). To control risk, Broadcom wants financially strong Microsoft to buy the first batch of 40% of the chips, install them in its own data centers, and then lease them back to OpenAI, using this as a guarantee to recoup the funding. If Microsoft does not buy, OpenAI will have to find someone else to take over. However, Microsoft has only set aside space in its data centers and has not made any purchase commitment. The OpenAI memo bluntly states that this has become “ongoing systemic risk” for the project.
OpenAI wants to pursue independence by making chips, but it is now tied down even more tightly. Sachin Katti, an executive in charge of the project, warned internally that treating Microsoft’s purchase commitment as a financing prerequisite would “seriously limit long-term development,” making the deal financially unattractive—yet the company still decided to move forward because of its strategic value. In addition, due to Broadcom’s tight allocation of capacity at TSMC, the first inference chip, codenamed Jalapeno, will not be rolled out at scale until 2027. As a concession, Broadcom has recently dropped its insistence on OpenAI having to contribute “one-for-one, equal amounts,” agreeing to shoulder more upfront funding.