{"id":15243,"date":"2026-06-19T05:17:37","date_gmt":"2026-06-19T05:17:37","guid":{"rendered":"https:\/\/wildgreenquest.com\/?p=15243"},"modified":"2026-06-19T05:17:37","modified_gmt":"2026-06-19T05:17:37","slug":"the-backbone-of-ai-unscrambling-the-basics","status":"publish","type":"post","link":"https:\/\/wildgreenquest.com\/?p=15243","title":{"rendered":"The Backbone Of AI: Unscrambling The Basics"},"content":{"rendered":"<p><br \/>\n<\/p>\n<div>\n<figure class=\"embed-base image-embed embed-0\" role=\"presentation\">\n<div style=\"padding-top:56.12%;position:relative\" class=\"image-embed__placeholder\"><picture><source media=\"(min-width: 960px)\" sizes=\"50vw\" srcset=\"https:\/\/imageio.forbes.com\/specials-images\/imageserve\/69fb7e9500cc0daf39ea9e53\/Curved-Data-Lines-With-Shallow-Depth-Of-Field---Blue---Technology--Computer-Network-\/0x0.jpg?width=960&amp;dpr=1 1x, https:\/\/imageio.forbes.com\/specials-images\/imageserve\/69fb7e9500cc0daf39ea9e53\/Curved-Data-Lines-With-Shallow-Depth-Of-Field---Blue---Technology--Computer-Network-\/0x0.jpg?width=960&amp;dpr=1.5 1.5x, https:\/\/imageio.forbes.com\/specials-images\/imageserve\/69fb7e9500cc0daf39ea9e53\/Curved-Data-Lines-With-Shallow-Depth-Of-Field---Blue---Technology--Computer-Network-\/0x0.jpg?width=960&amp;dpr=2 2x\"\/><\/picture><\/div>\n<\/figure>\n<p>You may or may not have heard this phrase before: \u201cthe backbone of AI.\u201d What does it mean? What is the \u201cbackbone\u201d of an AI system?<\/p>\n<p>I found this from Copilot:<\/p>\n<p>\u201cIn modern artificial intelligence, the backbone typically refers to the core neural network architecture responsible for extracting meaningful features from raw input data. This concept is especially common in computer vision, where a backbone is a pre-trained Convolutional Neural Network (CNN) used as the feature extractor for more complex tasks like object detection, segmentation, or image captioning.\u201d<\/p>\n<p>That\u2019s one way to describe it, and I\u2019ve often said that the convolutional neural network is central to AI, as it allows the entity to \u201csee\u201d what\u2019s around it, to start to process environmental data, which makes deployed AI an order of magnitude stronger. I don\u2019t know if I\u2019d call it the backbone, though.<\/p>\n<p>Here\u2019s something from a writer self-identified as \u201cPete the Momentum\u201d<a rel=\"nofollow\" href=\"https:\/\/petethemomentum.substack.com\/p\/coreweave-the-hidden-backbone-of\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" data-ga-track=\"ExternalLink:https:\/\/petethemomentum.substack.com\/p\/coreweave-the-hidden-backbone-of\" aria-label=\"identifying Coreweave systems\"> <u data-ga-track=\"ExternalLink:https:\/\/petethemomentum.substack.com\/p\/coreweave-the-hidden-backbone-of\">identifying Coreweave systems<\/u><\/a> as a \u201cbackbone\u201d for AI operators, though I can\u2019t tell if this is a puff piece or something else.<\/p>\n<p>I found this to be a little more helpful: a<a rel=\"nofollow\" href=\"https:\/\/www.datacenterknowledge.com\/networking\/building-the-backbone-of-ai-why-infrastructure-matters-in-the-race-for-adoption\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" data-ga-track=\"ExternalLink:https:\/\/www.datacenterknowledge.com\/networking\/building-the-backbone-of-ai-why-infrastructure-matters-in-the-race-for-adoption\" aria-label=\"Data Center Knowledge piece\"> <u data-ga-track=\"ExternalLink:https:\/\/www.datacenterknowledge.com\/networking\/building-the-backbone-of-ai-why-infrastructure-matters-in-the-race-for-adoption\">Data Center Knowledge piece<\/u><\/a> from Ivo Ivanov of Industry Perspectives, who cited our own MIT publication, writing:<\/p>\n<p>\u201cAccording to a 2024 <a rel=\"nofollow\" href=\"https:\/\/www.technologyreview.com\/2024\/08\/05\/1095447\/a-playbook-for-crafting-ai-strategy\/#:~:text=Fully%2095%25%20of%20companies%20surveyed,one%20to%20three%20use%20cases.\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" data-ga-track=\"ExternalLink:https:\/\/www.technologyreview.com\/2024\/08\/05\/1095447\/a-playbook-for-crafting-ai-strategy\/#:~:text=Fully%2095%25%20of%20companies%20surveyed,one%20to%20three%20use%20cases.\" aria-label=\"MIT Technology Review\"><u data-ga-track=\"ExternalLink:https:\/\/www.technologyreview.com\/2024\/08\/05\/1095447\/a-playbook-for-crafting-ai-strategy\/#:~:text=Fully%2095%25%20of%20companies%20surveyed,one%20to%20three%20use%20cases.\">MIT Technology Review<\/u><\/a> survey, a staggering 95% of businesses are already utilizing AI in some way, and more than half are aiming for full-scale integration in the next two years. The momentum behind AI is nothing short of remarkable, but as with any emerging technology, there are peaks and troughs before a state of blissful equilibrium is reached.\u201d<\/p>\n<section id=\"boston-discussion-imagination-action\">\n<h2 class=\"subhead-embed\">Boston Discussion at Imagination in Action<\/h2>\n<p>Then there\u2019s this from our April event at MIT from Imagination in Action, an organization I am affiliated with. My colleague Dave Blundin interviewed Jeremy Kepner, head of the MIT Lincoln Lab supercomputing center, Libby Wayman of Breakthrough Energy Ventures, and Chase Lockmiller of Crusoe, who zoomed in via teleconference.<\/p>\n<\/section>\n<section id=\"energy-equation\">\n<h2 class=\"subhead-embed\">The Energy Equation<\/h2>\n<p>\u201cI think AI is driving a multiple order of magnitude change in the amount of power requirement to run global computing infrastructure,\u201d Lockmiller said, elaborating on data center plans around the country, in places like Northern Virginia and Denver, Colorado. \u201cIt creates this huge, huge opportunity to create an abundance of not just intelligence, but also of energy.\u201d<\/p>\n<p>Keep in mind that Lockmiller\u2019s company is involved, as he mentioned, in the Abilene, Texas plant that is a part of the Operation Stargate project with a staggering $50 billion allocation, that was announced in the White House.<\/p>\n<p>Wayman, calling the AI arms race a \u201cU.S. and China phenomenon,\u201d spoke to some of the challenges involved in getting us to where we want to be.<\/p>\n<p>\u201cIn the United States, about 4.4% of U.S. electricity consumption went to data centers by the end of 2024, so zooming out from the specific data points that Chase shared, the trajectory that we&#8217;re on is that that data centers could consume about 20% of U.S. electricity, which is basically expanding the electric grid and electric consumption in the U.S. by 20%, by the end of this decade,\u201d she said. \u201cThis constraint that we&#8217;re feeling right now is really just from a couple percent of increase, and to take that up to 20% increase is going to put a massive strain on the system. What that&#8217;s looking like in individual local economies is: you&#8217;re starting to see some modeling that indicates that electricity prices will go up, emissions will go up.\u201d<\/p>\n<\/section>\n<section id=\"mit-contributions\">\n<h2 class=\"subhead-embed\">MIT Contributions<\/h2>\n<p>\u201cWe observed this energy importance in computing, I think, around 2002,\u201d Kepner said, \u201cand it was fairly obvious to us here at MIT, that energy is going to be very important.\u201d<\/p>\n<p>He talked about standing up a \u201cnext-gen\u201d data center in a former industrial town in Holyoke, Massachusetts, next to a hydroelectric plant.<\/p>\n<p>\u201cWe wanted it to be, not just great for computing, but we were really thinking long term,\u201d he continued. \u201cWe knew no one was going to buy us another one of these, and we could tell that the technology was going to be changing, this GPU technology was going to be changing, so we designed an extremely flexible data center, so that we could accommodate essentially any generation of technology.\u201d<\/p>\n<\/section>\n<section id=\"language-progress\">\n<h2 class=\"subhead-embed\">Language, and Progress<\/h2>\n<p>Kepner had two more interesting things to say: the first was around the industry of supercomputing, where most people assume that progress has been meteoric.<\/p>\n<p>\u201cPeople often ask me, like, hasn&#8217;t this changed so much over the years?\u201d he said. \u201cAnd I&#8217;ve been in supercomputing for a very long time, and as far as we&#8217;re concerned, nothing has changed. These computers do the exact same mathematical operation, which is called matrix multiply, that they&#8217;ve done for 40 years. It is the only operation we know how to accelerate. We don&#8217;t know how to accelerate anything else. And in fact, if AI could not take advantage of that operation, we are not talking about it. We are talking about something else. And so that has been the fundamental driver of this industry for four decades.<\/p>\n<p>Interesting.<\/p>\n<p>The other point is around where investors can put their backing and effort, as Kepner urged us to \u201cbet on math.\u201d<\/p>\n<p>\u201cMath is as much of a human language as English or French or anything, and the AIs understand it far better than English,\u201d he said.<\/p>\n<\/section>\n<section id=\"making-things-go\">\n<h2 class=\"subhead-embed\">Making Things Go<\/h2>\n<p>Toward the end of the talk, Lockmiller gave us a pretty granular view of how management happens around these cutting-edge projects, as speed becomes a top priority.<\/p>\n<p>\u201cLibby mentioned the speed of things,\u201d he said. \u201cYou know, when we were building out the initial phases for Abilene, I committed to us being able to do it in a year. And we were going through this process of trying to order a lot of different things from vendors.\u201d<\/p>\n<p>He cited something called a Power Distribution Center, which uses medium voltage power, doling it out to low voltage transformers.<\/p>\n<p>\u201cThe lead time of that was over 100 weeks,\u201d Lockmiller said, \u201cand this was from multiple different vendors. And I said, \u2018Okay, well, that doesn&#8217;t work for me, because I said I was going to get this done in a year.\u2019 So we actually had a facility where we&#8217;re manufacturing some components. And I turned to my team and I said, \u2018how quickly can we make this?\u2019 and we eventually got that done in 22 weeks.\u201d<\/p>\n<p>That\u2019s a little bit about what\u2019s going on at the top of some of these big projects. Stay tuned for more.<\/p>\n<\/section>\n<\/div>\n<p><br \/>\n<br \/><a href=\"https:\/\/www.forbes.com\/sites\/johnwerner\/2026\/06\/18\/the-backbone-of-ai-unscrambling-the-basics\/\">Source link <\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>You may or may not have heard this phrase before: \u201cthe backbone of AI.\u201d What does it mean? What is the \u201cbackbone\u201d of an AI system? I found this from Copilot: \u201cIn modern artificial intelligence, the backbone typically refers to the core neural network architecture responsible for extracting meaningful features from raw input data. This<\/p>\n","protected":false},"author":1,"featured_media":15244,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[37],"tags":[],"class_list":["post-15243","post","type-post","status-publish","format-standard","has-post-thumbnail","category-brand-spotlights"],"_links":{"self":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts\/15243","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=15243"}],"version-history":[{"count":0,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/posts\/15243\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=\/wp\/v2\/media\/15244"}],"wp:attachment":[{"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=15243"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=15243"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wildgreenquest.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=15243"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}