Google's TPU Rebuild: Why AI Agents Need 55x More Compute
Om avsnittet
Agent-driven traffic surged 7,800% in one year, and serving projected 2030 demand takes a 55-fold leap in compute efficiency. See how Google is rebuilding its stack from TPU silicon to networking and memory, and why that changes the economics of AI infrastructure.
• DeepValues app, a research team in your pocket: intrinsic value and margin of safety for any stock, reports down to the footnotes
• Is the market overvalued? The Buffett Indicator app
Android (Buffett Indicator & Shiller PE)
Chapters
(0:00) Agentic AI Architecture
(1:16) From Queries to Agentic Loops
(2:16) Compute Intensity Surge
(3:57) From Queries to Agentic Loops
(4:56) Tensor Processing Unit Anatomy
(7:00) Smarter Network Topologies
(8:25) Tensor Processing Unit Anatomy
(9:22) Beating the Memory Wall
(10:04) Smarter Network Topologies
(12:02) Beating the Memory Wall
(12:43) 8th Gen TPU Family
(13:25) Beating the Memory Wall
(14:41) Open Software Frameworks
(15:20) Stateful MicroVMs
(19:23) Al Hypercomputer Ecosystem
(20:59) Project Suncatcher
(22:41) Al Hypercomputer Ecosystem
(23:40) Intelligent Investor Tools
(24:16) Get the apps
DeepValues toolkit
• Run the numbers on any stock yourself, free
• Free Margin of Safety Weekly, every Saturday (guru moves, market temperature, the week's research; no account needed)
• Open-source research on 300+ companies
• Deep Value Intelligence, the multi-agent AI analyst
• Developer API and MCP server for your AI agents
Deep Values explains AI from first principles: the research papers, models, chips and companies shaping the AI era, and what they mean for builders and investors. Understand AI from first principles.
Podcast (Deep Values Frontier): Spotify · Apple Podcasts
The two hosts' voices are AI-generated from the sources discussed. Educational content only; not investment advice.
#AI #ArtificialIntelligence #LLM
Original source
Deep Values Frontier: AI Research Explained med DeepValues finns tillgänglig på flera plattformar. Informationen på denna sida kommer från offentliga podd-flöden.