[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"article-ai-infrastructure-trillion-dollar-asset-class-en":3,"article-related-ai-infrastructure-trillion-dollar-asset-class-en":33,"series-industry-11b31146-7206-42ab-8f9f-da9cf0d98714":80},{"id":4,"slug":5,"title":6,"content":7,"summary":8,"source":9,"source_url":10,"author":11,"image_url":12,"cover_image":12,"category":13,"language":14,"translated_content":11,"related_article_id":15,"keywords":16,"key_takeaways":25,"views":29,"created_at":30,"published_at":31,"topic_cluster_id":32},"11b31146-7206-42ab-8f9f-da9cf0d98714","ai-infrastructure-trillion-dollar-asset-class-en","AI infrastructure is becoming a trillion-dollar asset class","\u003Cp data-speakable=\"summary\">\u003Ca href=\"\u002Ftag\u002Fai-infrastructure\">AI infrastructure\u003C\u002Fa> is now a trillion-dollar investment story, not just a software story.\u003C\u002Fp>\u003Cp>The numbers aren’t incremental. Goldman Sachs projects about $7.6 trillion in cumulative AI infrastructure capital expenditure from 2026 to 2031, covering compute, \u003Ca href=\"\u002Fnews\u002Fai-demand-starts-paying-for-data-centers-en\">data centers\u003C\u002Fa>, and power. That scale changes how investors, utilities, and policymakers should think about the sector.\u003C\u002Fp>\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Item\u003C\u002Fth>\u003Cth>What it covers\u003C\u002Fth>\u003Cth>Why it matters\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>Compute\u003C\u002Ftd>\u003Ctd>AI chips, servers, networking\u003C\u002Ftd>\u003Ctd>Drives model training and inference capacity\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Data centers\u003C\u002Ftd>\u003Ctd>Buildings, cooling, racks, site buildout\u003C\u002Ftd>\u003Ctd>Determines where capacity can be deployed\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>Power\u003C\u002Ftd>\u003Ctd>Grid connections, generation, transmission\u003C\u002Ftd>\u003Ctd>Sets the ceiling on how fast AI can scale\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003Ch2>1. Compute is now a capital market story\u003C\u002Fh2>\u003Cp>AI demand is no longer just about better models. It is about the physical stock of chips, servers, and networking gear needed to train and run those models at scale. Once spending reaches trillions, compute starts to look like a strategic industrial asset rather than a narrow IT purchase.\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782826367545-pqf5.png\" alt=\"AI infrastructure is becoming a trillion-dollar asset class\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>That shift matters because compute supply is constrained by manufacturing capacity, export controls, and the pace of hardware refreshes. Investors and policymakers need to treat it as a long-duration bet with real bottlenecks.\u003C\u002Fp>\u003Cul>\u003Cli>AI accelerators and GPUs\u003C\u002Fli>\u003Cli>High-speed networking equipment\u003C\u002Fli>\u003Cli>Server racks and storage systems\u003C\u002Fli>\u003Cli>Replacement cycles tied to model growth\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>2. Data centers are the physical backbone\u003C\u002Fh2>\u003Cp>Data centers are where AI capital becomes operational capacity. They need land, permits, cooling, fiber access, and local support before a single model can run. As demand rises, the bottleneck is often not the machine itself but the building that houses it.\u003C\u002Fp>\u003Cp>This is why site selection, zoning, and construction timelines matter as much as chip procurement. A data center delayed by power access or permitting can slow deployment even when demand is ready to go.\u003C\u002Fp>\u003Cul>\u003Cli>Site acquisition and entitlement\u003C\u002Fli>\u003Cli>Cooling systems and water use\u003C\u002Fli>\u003Cli>Fiber and network connectivity\u003C\u002Fli>\u003Cli>Construction lead times and permitting\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>3. Power is the binding constraint\u003C\u002Fh2>\u003Cp>AI infrastructure cannot grow faster than the grid can support it. Large-scale compute clusters draw serious electricity, and the forecasted spending wave implies much higher demand for generation, transmission, and interconnection capacity.\u003C\u002Fp>\n\u003Cfigure class=\"my-6\">\u003Cimg src=\"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782826366259-k6sz.png\" alt=\"AI infrastructure is becoming a trillion-dollar asset class\" class=\"rounded-xl w-full\" loading=\"lazy\" \u002F>\u003C\u002Ffigure>\n\u003Cp>That creates a policy problem as much as an engineering one. If Washington wants AI expansion to happen in the United States, it has to make it easier to build power assets and connect them to load centers.\u003C\u002Fp>\u003Cul>\u003Cli>New generation capacity\u003C\u002Fli>\u003Cli>Transmission buildout\u003C\u002Fli>\u003Cli>Interconnection queues\u003C\u002Fli>\u003Cli>Grid reliability and load planning\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>4. Policy now affects investment timing\u003C\u002Fh2>\u003Cp>When a sector reaches this scale, regulation stops being background noise. Permitting rules, energy policy, tax treatment, and trade restrictions all shape where capital flows and how quickly projects move from announcement to operation. That is especially true for AI, where the physical stack is tightly linked to national policy.\u003C\u002Fp>\u003Cp>Washington’s job is not to pick winners. It is to remove avoidable delays and make sure the rules around power, land, and hardware are aligned with the pace of private investment.\u003C\u002Fp>\u003Cul>\u003Cli>Permitting reform\u003C\u002Fli>\u003Cli>Grid modernization incentives\u003C\u002Fli>\u003Cli>Tax policy for capital formation\u003C\u002Fli>\u003Cli>Trade and export rules for AI hardware\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>5. Capital allocation will favor the full stack\u003C\u002Fh2>\u003Cp>The $7.6 trillion estimate covers more than servers. It points to a broader investment thesis across compute, buildings, and electricity. That means the best-positioned firms may be those that sit across the full stack, from chip suppliers to utilities to \u003Ca href=\"\u002Fnews\u002Fe-estate-leads-2026-real-estate-tokenization-en\">real estate\u003C\u002Fa> and infrastructure developers.\u003C\u002Fp>\u003Cp>For readers tracking the opportunity, the key insight is that AI spending is becoming an infrastructure cycle. The winners will not all be software companies; many will be the firms that build and power the system underneath the software.\u003C\u002Fp>\u003Cul>\u003Cli>Chipmakers and server vendors\u003C\u002Fli>\u003Cli>Data center operators\u003C\u002Fli>\u003Cli>Utilities and grid developers\u003C\u002Fli>\u003Cli>Construction and engineering firms\u003C\u002Fli>\u003C\u002Ful>\u003Ch2>How to decide\u003C\u002Fh2>\u003Cp>If you are an investor, focus on companies tied to compute supply, data center buildout, and power delivery. If you are a policymaker, focus on permitting, grid capacity, and the rules that slow private capital. If you are a business leader, plan for higher infrastructure costs and longer deployment timelines.\u003C\u002Fp>\u003Cp>The core takeaway is simple: AI is becoming an asset class with real-world constraints, and the binding constraint is no longer code. It is steel, silicon, and electricity.\u003C\u002Fp>","1 trillion-dollar forecast explains why AI infrastructure now demands policy, capital, and power planning.","www.aei.org","https:\u002F\u002Fwww.aei.org\u002Ftechnology-and-innovation\u002Fai-infrastructure-is-a-new-asset-class-and-washington-has-work-to-do\u002F",null,"https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782826367545-pqf5.png","industry","en","99d12162-7c57-4d83-a454-374e8c832c01",[17,18,19,20,21,22,23,24],"AI infrastructure","data centers","compute","power grid","capital expenditure","Goldman Sachs","Washington policy","AI chips",[26,27,28],"Goldman Sachs projects about $7.6 trillion in AI infrastructure capex from 2026 to 2031.","Compute, data centers, and power are the three main bottlenecks shaping AI expansion.","Policy on permitting, grids, and hardware trade will affect where AI capital goes.",0,"2026-06-30T13:32:23.803581+00:00","2026-06-30T13:32:23.793+00:00","50ad070c-8891-4ccc-a7ee-038aa8918c86",{"tags":34,"relatedLang":39,"relatedPosts":43},[35,37],{"name":18,"slug":36},"data-centers",{"name":17,"slug":38},"ai-infrastructure",{"id":15,"slug":40,"title":41,"language":42},"ai-infrastructure-trillion-dollar-asset-class-zh","AI 基礎設施正變成兆級資產","zh",[44,50,56,62,68,74],{"id":45,"slug":46,"title":47,"cover_image":48,"image_url":48,"created_at":49,"category":13},"1f51043a-eaa0-47b2-ba3a-600d525fdbf2","free-ai-model-picks-that-actually-run-today-en","Free AI model picks that actually run today","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782837189810-lzqu.png","2026-06-30T16:32:31.559964+00:00",{"id":51,"slug":52,"title":53,"cover_image":54,"image_url":54,"created_at":55,"category":13},"790789d3-c794-430e-93b8-20614574dea1","ai-demand-starts-paying-for-data-centers-en","AI Demand Starts Paying for Data Centers","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782825479139-dy7a.png","2026-06-30T13:17:34.991501+00:00",{"id":57,"slug":58,"title":59,"cover_image":60,"image_url":60,"created_at":61,"category":13},"bc3a5cde-08c3-45db-89c7-1ca4419c1d4e","ai-should-govern-sdlc-before-code-en","AI should govern the SDLC before it writes code","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782822779071-04gq.png","2026-06-30T12:32:21.899561+00:00",{"id":63,"slug":64,"title":65,"cover_image":66,"image_url":66,"created_at":67,"category":13},"952734bd-be94-4603-8606-810ff75c4be2","rwa-tokenization-new-default-ownership-2026-en","RWA tokenization is becoming the new default for ownership","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782819166578-u7f2.png","2026-06-30T11:32:22.166351+00:00",{"id":69,"slug":70,"title":71,"cover_image":72,"image_url":72,"created_at":73,"category":13},"3f89e362-9464-415b-8522-202e8189e004","e-estate-leads-2026-real-estate-tokenization-en","E-Estate leads 2026 real estate tokenization","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782818270114-lhqn.png","2026-06-30T11:17:26.585416+00:00",{"id":75,"slug":76,"title":77,"cover_image":78,"image_url":78,"created_at":79,"category":13},"3249ec56-0940-4cae-a21b-053e899ba56e","docker-monitoring-tools-real-budgets-en","12 Docker monitoring tools that fit real budgets","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1782806590248-vm2p.png","2026-06-30T08:02:42.276141+00:00",[81,86,91,96,101,106,111,116,121,126],{"id":82,"slug":83,"title":84,"created_at":85},"d35a1bd9-e709-412e-a2df-392df1dc572a","ai-impact-2026-developments-market-en","AI's Impact in 2026: Key Developments and Market Shifts","2026-03-25T16:20:33.205823+00:00",{"id":87,"slug":88,"title":89,"created_at":90},"5ed27921-5fd6-492e-8c59-78393bf37710","trumps-ai-legislative-framework-en","Trump's AI Legislative Framework: What's Inside?","2026-03-25T16:22:20.005325+00:00",{"id":92,"slug":93,"title":94,"created_at":95},"e454a642-f03c-4794-b185-5f651aebbaca","nvidia-gtc-2026-key-highlights-innovations-en","NVIDIA GTC 2026: Key Highlights and Innovations","2026-03-25T16:22:47.882615+00:00",{"id":97,"slug":98,"title":99,"created_at":100},"0ebb5b16-774a-4922-945d-5f2ce1df5a6d","claude-usage-diversifies-learning-curves-en","Claude Usage Diversifies, Learning Curves Emerge","2026-03-25T16:25:50.770376+00:00",{"id":102,"slug":103,"title":104,"created_at":105},"69934e86-2fc5-4280-8223-7b917a48ace8","openclaw-ai-commoditization-concerns-en","OpenClaw's Rise Raises Concerns of AI Model Commoditization","2026-03-25T16:26:30.582047+00:00",{"id":107,"slug":108,"title":109,"created_at":110},"b4b2575b-2ac8-46b2-b90e-ab1d7c060797","google-gemini-ai-rollout-2026-en","Google's Gemini AI Rollout Extended to 2026","2026-03-25T16:28:14.808842+00:00",{"id":112,"slug":113,"title":114,"created_at":115},"6e18bc65-42ae-4ad0-b564-67d7f66b979e","meta-llama4-fabricated-results-scandal-en","Meta's Llama 4 Scandal: Fabricated AI Test Results Unveiled","2026-03-25T16:29:15.482836+00:00",{"id":117,"slug":118,"title":119,"created_at":120},"bf888e9d-08be-4f47-996c-7b24b5ab3500","accenture-mistral-ai-deployment-en","Accenture and Mistral AI Team Up for AI Deployment","2026-03-25T16:31:01.894655+00:00",{"id":122,"slug":123,"title":124,"created_at":125},"5382b536-fad2-49c6-ac85-9eb2bae49f35","mistral-ai-high-stakes-2026-en","Mistral AI: Facing High Stakes in 2026","2026-03-25T16:31:39.941974+00:00",{"id":127,"slug":128,"title":129,"created_at":130},"9da3d2d6-b669-4971-ba1d-17fdb3548ed5","cursors-meteoric-rise-pressures-en","Cursor's Meteoric Rise Faces Industry Pressures","2026-03-25T16:32:21.899217+00:00"]