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 Mobile Supply Chain: Ming-Chi Kuo Predicts MediaTek Will Dominate Processors, Mass Production Expected as Early as 2027
Analysts point out that OpenAI is accelerating the development of AI phones, with mass production expected as early as the first half of 2027. MediaTek is expected to secure exclusive orders for customized chips.
Hong Kong Tianfeng International Securities analyst Guo Mingqi only announced on April 27 that OpenAI, the parent company of ChatGPT, is collaborating with MediaTek and Qualcomm to develop a smartphone processor, with Luxshare Precision serving as the exclusive system design and manufacturing partner, with a planned mass production date as early as 2028. This sparked a reimagining in the market: OpenAI’s hardware strategy may not just be about creating a device equipped with ChatGPT, but about redefining smartphones with an “AI Agent Phone.”
By May 5, Guo Mingqi further added that the latest industry survey describes a more aggressive development progress for OpenAI’s first AI Agent phone. He pointed out that OpenAI might be accelerating the development of its first AI Agent phone, with a target of mass production as early as the first half of 2027. This suggests that the latest schedule could be about six months to a year earlier than the previously mentioned “mass production by 2028” on April 27.
Guo Mingqi believes that the reason for OpenAI’s accelerated phone development could include facilitating its end-of-year IPO narrative and the increasing competition in AI Agent phones. In other words, if OpenAI can clearly demonstrate its hardware strategy before the IPO, it can not only strengthen its growth story around “AI operating systems” and “AI personal agent portals,” but also show the capital markets that it is not just a cloud-based model company, but one with the potential to enter the next-generation consumer computing platform.
From joint development with Qualcomm and MediaTek to MediaTek potentially securing exclusive processor orders
In the April 27 version, Guo Mingqi mentioned that OpenAI is collaborating with MediaTek and Qualcomm to develop smartphone processors, with the supply chain still showing a competitive or co-developing landscape.
However, the May 5 update shows a clearer shift: Guo Mingqi indicated that MediaTek is now more likely to secure exclusive orders for the OpenAI AI Agent phone processor. The device is expected to use a customized version based on Dimensity 9600, manufactured by TSMC’s N2P process in the second half of 2026.
This shift is quite significant. If MediaTek ultimately wins the processor order exclusively, it would mean that it is not just a participant in the OpenAI phone supply chain, but could become the core chip platform provider for the first-generation AI Agent phone. For MediaTek, which has long been chasing Qualcomm in the high-end mobile chip market, making the OpenAI phone a new benchmark for AI devices could be a major revaluation of its brand and technological standing.
Especially since the focus of the OpenAI phone will not just be on traditional CPUs, GPUs, or communication specs, but on how to support AI Agents in understanding contexts, processing tasks, managing memory, and scheduling cloud models over long periods. This greatly increases the importance of customized processors.
Key specifications: ISP, dual NPU, LPDDR6, UFS 5.0, and security architecture
In the May 5 update, Guo Mingqi also provided more specific details on the specifications. He pointed out that the customized chip for the OpenAI AI Agent phone will enhance the ISP’s high dynamic range output, which is beneficial for real-world visual perception, making it a key specification.
This also aligns with the core logic of the AI Agent phone. If the phone is to serve as an entry point for AI agents, it cannot rely solely on text input but must understand the user’s environment, images, scenes, and real-time states. Cameras and sensors are no longer just tools for taking photos but are vital data sources for AI Agents to understand the real world. Therefore, high dynamic range output from the ISP may be designed to ensure stable visual perception in complex lighting, indoor-outdoor switching, and moving scenes.
Other key specifications include a dual NPU architecture, LPDDR6 paired with UFS 5.0, and security features like pKVM with inline hashing.
Approximately 30 million units expected to be shipped in 2027 and 2028 combined
Guo Mingqi also provided a clearer shipment forecast. If development proceeds smoothly, the OpenAI AI Agent phone is expected to ship around 30 million units combined in 2027 and 2028.
While this number still lags significantly behind the annual iPhone shipment volume, it is a very ambitious target for a completely new brand and a new form of AI phone. If OpenAI can mass produce in the first half of 2027, it suggests that the device is not just an experimental hardware or niche developer device, but an attempt to enter the mainstream consumer smartphone market.
This also makes the supply chain assessment on April 27 more meaningful for investment. Guo Mingqi pointed out that if OpenAI initially targets the global high-end smartphone market, which has an annual shipment scale of about 300 to 400 million units; now, if the first-generation product can reach 30 million units in 2027 to 2028, it indicates that OpenAI at least hopes to make the AI Agent phone a scalable new product category, rather than just a concept for AI hardware.