A research firm that began as a personal blog and now has the power to crash U.S. stocks with a single report—SemiAnalysis—and its non-semiconductor-background founder, @dylan522p: 1. Dylan Patel was born in 1996 and holds a Bachelor’s degree in Management and Law from the University of Georgia. From the age of 8 to 12, he was already active on online hardware forums, teaching himself hardware repair and chip technology as a “forum warrior” by reading documentation and engaging in community discussions. For years, he anonymously published chip and hardware analyses on Reddit, WordPress blogs, and Silicon Twitter, while moderating multiple hardware communities including those for Nvidia. 2. On May 22, 2020—his 24th birthday—Dylan published his first blog post and officially founded SemiAnalysis. It began as an anonymous personal WordPress blog before transitioning into a paid Substack newsletter. Initially, Dylan worked alone, focusing exclusively on in-depth analysis of semiconductor supply chains, AI infrastructure, cloud ecosystems, machine learning models, and related fields. Over time, he expanded into consulting services. Dylan himself evolved from an anonymous “Silicon Twitter chip blogger” into one of the most closely followed AI infrastructure analysts globally. 3. In recent years, SemiAnalysis has grown rapidly and is now among the world’s leading research institutions in AI infrastructure and semiconductors, with a team of approximately 60 people distributed across 8–10 countries including the U.S., Japan, Taiwan, Singapore, France, Germany, Israel, Canada, and the UK. SemiAnalysis is projected to generate over $100 million in revenue by 2026 (per The Information), representing explosive growth from its early subscription model. Its business model includes: 1) Paid subscriptions (newsletter + in-depth reports); 2) Proprietary industry models (e.g., AI Datacenter Model, Accelerator Industry Model, Wafer Fab Model, AI Cloud TCO Model, AI Networking Model) used to forecast wafer capacity, data center power consumption, accelerator output, and TCO economics; 3) Consulting services, custom projects, and hourly advisory. It provides a unified view across the entire supply chain—from fabs to the cloud to ML models. These proprietary models and datasets (such as satellite imagery monitoring data center construction, GitHub commit analysis, and HBM supply chain tracking) have become critical decision-making tools for clients including top AI labs, hyperscalers, hedge funds, and semiconductor giants. Its analyses directly influence trillions of dollars in AI capital expenditure decisions. 4. Even more critically, SemiAnalysis now wields extraordinary influence: 1) At this year’s GTC keynote, Jensen Huang publicly named Dylan Patel, displaying SemiAnalysis’s latest InferenceX chip performance report on screen and dedicating five full minutes to analyzing it—then continued referencing it during Q&A. For a figure like Jensen to personally call out an independent analyst carries immense weight. 2) Lisa Su went even further: After SemiAnalysis published a scathing report criticizing AMD’s MI300X training performance, Lisa Su personally called Dylan the next day to schedule a 90-minute one-on-one meeting—and later publicly tweeted: “Feedback is a gift even when it’s critical.” This move effectively elevated an independent research firm to CEO-level strategic dialogue. From this perspective, SemiAnalysis is no longer merely a “newsletter”—it has become the strategic intelligence hub of the AI semiconductor ecosystem. Its influence is amplified threefold through proprietary data models, frequent media exposure, and dual-sided client engagement—directly shaping engineering decisions in Silicon Valley and multi-billion-dollar trades on Wall Street. Today, hedge funds and asset managers routinely rely on SemiAnalysis’s monthly ChipBook and proprietary datasets as standard practice—for idea generation, thesis validation, and portfolio tracking. A single report can trigger immediate stock price movements and adjustments in hundreds of billions of capex. Hyperscalers planning custom ASICs, GPU procurement, or data center power allocation all consult its supply chain breakdowns. It is no longer merely a “reference”—it directly dictates decisions. This is a classic case of “non-traditional 0 to 1”: no elite EE PhD from a top university, no background at a major chip company—just self-taught expertise, data-driven analysis, and independent thinking. From an anonymous blog to earning the serious attention of Nvidia’s and AMD’s CEOs. Silicon Valley and Wall Street now assume: if you want to understand the real pace of AI hardware development, first check what SemiAnalysis and Dylan have to say. This is arguably the most extreme possible manifestation of independent investment research by a single individual.