@MrPingChecks: This is my favorite chip expert, who made very accurate bets on tech this year. On March 25, he wrote an article strong…
Summary
据称 Google TPU 首席设计工程师透露，AI 已接管其大部分芯片设计工作，且内部模型比公开版 Gemini 强大得多，引发对 AI 驱动芯片设计趋势的关注。
View Cached Full Text
Cached at: 10/11/26, 03:17 PM
This is my favorite chip expert, who made very accurate bets on tech this year. On March 25, he wrote an article strongly recommending storage, and soon after, MU tripled. At the end of July, when optical modules plummeted, he said he bought Lumentum, which then rose another 47%. Today he dropped another bombshell: Google’s TPU chief design engineer told him that AI has already taken over most chip design, using an internal model that’s far more powerful than the public version of Gemini.
Damnang (@damnang2): I met a lead design engineer working on Google’s TPUs. He told me that AI has already taken over most of their design work and does it more efficiently than humans.
He said they use their own internal model, which is much smarter than the publicly available Gemini, and they’re
Similar Articles
Can Google’s AI Chip Beat Nvidia?
A long-form tech deep dive analyzing whether Google's TPU chips can challenge Nvidia's dominance in GPUs — arguing that the outcome hinges not only on the hardware itself, but on the CUDA software ecosystem, cost economics, and the way large customers like Anthropic are spreading their bets across both types of chips.
Google is working on a new AI chip designed to make Gemini more efficient
Google is designing a new AI server chip, Frozen v2, to improve the efficiency of its Gemini models, aiming for 6-10x improvement over existing chips, with release targeted for 2028.
Google reportedly taps AMD to design next-generation TPU — hybrid AI ASIC could integrate on-package CPU cores for reinforcement learning (3 minute read)
Google is reportedly collaborating with AMD to design a next-generation TPU that could integrate CPU cores on-package for reinforcement learning workloads, potentially advancing AI hardware architecture.
Google just unveiled its newest AI chips
Google unveiled eighth-gen TPUs (8t/8i) and a new Gemini Enterprise Agent Platform at Cloud Next, while revealing 75% of new Google code is now AI-generated.
@snowboat84: https://x.com/snowboat84/status/2061962883651731602
This article is the first part of the AI Engineering Panorama series. From a historical perspective, it reviews the evolution of GPUs from gaming graphics cards to AI accelerators, the bold bet of CUDA, the independent path of Google's TPU, and why NVIDIA ultimately prevailed. It also provides a detailed analysis of the underlying logic of AI infrastructure such as chips, supply chain, networking, and power.