<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Enes Karataş on Medium]]></title>
        <description><![CDATA[Stories by Enes Karataş on Medium]]></description>
        <link>https://medium.com/@enes.onion?source=rss-2258ab147828------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*VPD242WNVnQvjYbgl1RY5Q.jpeg</url>
            <title>Stories by Enes Karataş on Medium</title>
            <link>https://medium.com/@enes.onion?source=rss-2258ab147828------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 04:33:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@enes.onion/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Doğadaki Cevheri Keşfet: Türkiye’nin İlk Kırıntı Madencilik Topluluğu Mobil Uygulaması Yayında!]]></title>
            <link>https://medium.com/@enes.onion/do%C4%9Fadaki-cevheri-ke%C5%9Ffet-t%C3%BCrkiyenin-i%CC%87lk-k%C4%B1r%C4%B1nt%C4%B1-madencilik-toplulu%C4%9Fu-mobil-uygulamas%C4%B1-yay%C4%B1nda-0422cd385c77?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/0422cd385c77</guid>
            <category><![CDATA[altın-arama]]></category>
            <category><![CDATA[mobil-uygulama]]></category>
            <category><![CDATA[kırıntı-madencilik]]></category>
            <category><![CDATA[altın]]></category>
            <category><![CDATA[doğal-taşlar]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Mon, 27 Jul 2026 06:55:42 GMT</pubDate>
            <atom:updated>2026-07-27T06:55:42.298Z</atom:updated>
            <content:encoded><![CDATA[<p>Doğada bulduğunuz o parlak taş gerçekten altın mı yoksa pirit (ahmak altını) mi? Kırıntı madenciliği serüvenine adım atmak için ihtiyacınız olan rehber ve topluluk artık cebinizde.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QA0iYx1kePToVJS-UTrUqQ.png" /><figcaption>Kırıntı Madencilik Topluluğu — Google Play Store Görseli</figcaption></figure><p>(Uygulamayı <a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura"><strong>Google Play Store</strong></a> üzerinden <a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura"><strong>indirmek</strong></a> için tıkla.)</p><p>Bir dere kenarında yürürken veya doğada kamp yaparken gözünüze çarpan o ilginç taşı elinize aldığınızı hayal edin. Rengi, parıltısı ve dokusuyla sıradan bir çakıl taşından çok daha fazlası gibi duruyor. Peki, o an aklınızdan geçen ilk soru ne olurdu?</p><p><em>“Acaba bu değerli bir maden mi?”</em></p><p>Kırıntı madenciliği (placer mining), doğayla iç içe olmayı seven, keşfetme duygusunu kaybetmemiş herkes için harika bir tutku. Ancak Türkiye’de bu hobiye ilgi duyanların ortak ve çok büyük bir sorunu vardı: <strong>Bilgi kirliliği, yalnızlık ve buldukları mineralleri analiz edememek.</strong></p><p>İşte <strong>Kırıntı Madencilik</strong> uygulaması tam olarak bu sorunu çözmek için doğdu.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YrBR1yZDgiwWKAVMCSgg-w.png" /><figcaption>Türkiye’nin ilk kırıntı madencilik mobil uygulaması</figcaption></figure><h3>Bir İhtiyaçtan Doğan Fikir: Nereden Başlayacağız?</h3><p>Kırıntı madenciliğine merak saldığınızda, karşınıza aşılması gereken birkaç engel çıkar. Ekipmanları nereden bulacağınızı, eleme işlemini (panning) nasıl yapacağınızı veya hangi derelerde arama yapmanın daha mantıklı olduğunu öğrenmek istersiniz. İnternette araştırma yaptığınızda ise genellikle yabancı kaynaklara veya kopuk forum sayfalarına denk gelirsiniz.</p><p>Geliştirici şapkamı takıp bu alana baktığımda, yerel bir topluluğun eksikliğini çok net hissettim. İnsanların buldukları mineralleri paylaşabilecekleri, tecrübeli isimlerden tavsiye alabilecekleri, kendisiyle aynı hobiye sahip farklı insanlarla birlikte ortak etkinlik yapabilme ve en önemlisi ceplerinde taşıyabilecekleri dijital bir rehbere ihtiyaçları vardı.</p><p>Bu vizyonla yola çıkarak, Türkiye’nin ilk kırıntı madenciliği topluluk ve asistan uygulamasını geliştirdim.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/682/1*bm7ShungYCAfi_5Luf-PCw.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü - 1</figcaption></figure><h3>Cebinizdeki Jeolog: Yapay Zeka ile Mineral Tanımlama</h3><p>Yeni başlayanların hevesini kıran en büyük şey, saatlerce uğraşıp buldukları taşın ne olduğunu bilememektir. Geleneksel yöntemlerle bir taşı analiz etmek laboratuvar ortamı veya ciddi bir jeoloji bilgisi gerektirir.</p><p>Uygulamayı geliştirirken en çok odaklandığım ve heyecan duyduğum özellik bu oldu. Kırıntı Madencilik uygulaması, entegre <strong>Yapay Zeka Destekli Mineral Tanımlama</strong> teknolojisiyle çalışıyor.</p><ul><li><strong>Nasıl Çalışıyor?</strong> Doğada bulduğunuz taşın veya mineralin fotoğrafını uygulama üzerinden çekiyorsunuz.</li><li><strong>Sonuç:</strong> Yapay zeka saniyeler içinde görseli analiz ederek size o taşın ne olabileceğine dair yüksek doğruluk oranına sahip tahminler ve maden hakkında teknik bilgiler sunuyor.</li><li>İsterseniz de bir arazi fotoğrafı da yükleyerek şüphelendiğiniz noktalar hakkında bir analiz yaptırtabilirsiniz.</li></ul><p>Artık “Bu bulduğum kuvars mı, yoksa değerli bir maden mi?” ya da “Acaba buradan altın çıkabilir mi?” sorularının cevabı sadece bir deklanşör uzaklığında.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/684/1*R_aufkp3zKN4lC_k8pqvhQ.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü — 2</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/686/1*ksq5llwipPzqwu7ofMahEA.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü — 3</figcaption></figure><h3>Sadece Bir Araç Değil, Bir Topluluk</h3><p>Madencilik, doğası gereği deneyim aktarımına dayanan bir alandır. Kırıntı Madencilik, bir mineral tanıma aracından çok daha fazlası. Uygulama içerisindeki topluluk özelliği sayesinde:</p><ul><li><strong>Keşiflerinizi Paylaşın:</strong> Eleme tavasında bulduğunuz ilk altın tozunu veya ilginç kuvars kristallerini diğer kullanıcılarla paylaşabilirsiniz.</li><li><strong>Ustalara Danışın:</strong> Emin olamadığınız konularda topluluktaki diğer madencilere sorular sorabilir, fikir alışverişinde bulunabilirsiniz.</li><li><strong>Birlikte Öğrenin:</strong> Ekipman tavsiyelerinden lokasyon okumaya kadar birçok konuda Türkçe ve güvenilir bir bilgi ağına anında erişebilirsiniz.</li></ul><h3>“Nerede Arayacağız?” İnteraktif Harita ve Arazi Analizleri</h3><p>Yapay zeka ile bulduğunuz taşı tanımlamak harika, peki ama aramaya nereden başlayacaksınız? Altın ve değerli mineraller rastgele dağılmazlar; suyun akış hızının düştüğü nehir kıvrımlarında, büyük kayaların arkasında veya ana kaya (bedrock) çatlaklarında birikirler. Kısacası doğayı ve topografyayı “okumak” gerekir.</p><p>İşte bu noktada uygulamanın <strong>İnteraktif Harita ve Arazi Analizleri</strong> özelliği devreye giriyor:</p><ul><li><strong>Jeolojik Okuma:</strong> Türkiye’nin zengin coğrafyasında potansiyel kırıntı madenciliği noktalarını, akarsu yataklarını ve jeolojik yapıları harita üzerinden inceleyebilirsiniz.</li><li><strong>Arazi Analizi:</strong> Hangi bölgelerin maden birikimi için daha elverişli olduğunu dijital ortamda analiz ederek, sahaya çıkmadan önce nokta atışı lokasyonlar belirleyebilirsiniz.</li><li><strong>Kişisel Noktalarınızı İşaretleyin:</strong> Kendi keşfettiğiniz verimli noktaları haritaya gizli olarak kaydedebilir, bir sonraki sefer için rotanızı önceden planlayabilirsiniz.</li></ul><p>Böylece doğada kör uçuşu yapmak yerine, elinizde güçlü bir dijital pusulayla yola çıkmış olursunuz.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/686/1*wTXZUbpcAZAwX4besp1Wcg.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü — 4</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/686/1*ftY_M-ei1amtezOkIRt0oA.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü — 5</figcaption></figure><h3>Bulduklarınızı Değerlendirin ve Eksiklerinizi Tamamlayın: Topluluk Pazaryeri</h3><p>Kırıntı madenciliği yaparken zamanla elinizde harika bir doğal taş koleksiyonu birikecek. Ya da belki de işleri biraz daha büyütmek isteyip yeni bir eleme tavasına (pan) veya savak (sluice box) sistemine ihtiyaç duyacaksınız. Peki, bu niş alandaki ihtiyaçlarınızı nereden karşılayacak veya bulduğunuz değerli taşları nerede değerlendireceksiniz?</p><p>Uygulamaya entegre ettiğimiz <strong>Pazaryeri</strong> özelliği tam olarak bu döngüyü sağlamak için var.</p><ul><li><strong>Doğal Taşlarınızı Satışa Çıkarın:</strong> Doğada bulduğunuz göz alıcı kuvarsları, ametistleri veya diğer doğal taşları, doğrudan bu alana ilgi duyan alıcılara ve koleksiyonerlere sergileyip satabilirsiniz.</li><li><strong>Ekipman Alışverişi Yapın:</strong> Hobiye yeni başlarken ikinci el, uygun fiyatlı bir başlangıç seti mi arıyorsunuz? Ya da elinizdeki eski ekipmanı satıp bütçe mi oluşturmak istiyorsunuz? İhtiyacınız olan tüm donanımlara, güvenilir madencilik topluluğunun içinden doğrudan ulaşabilirsiniz.</li></ul><p>Artık “Bu taşı kime satabilirim?” veya “Doğru ekipmanı nereden bulurum?” diye düşünmenize gerek yok. Hem alıcı hem de satıcı olarak kendi pazarınızı bu topluluğun içinde yaratabilirsiniz.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/682/1*gak0oU8gP_EYHGjGPmdzPA.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü — 6</figcaption></figure><h3>Akademi ve Bilgi Paylaşım Alanı: Kendi Dijital Kütüphanemizi İnşa Ediyoruz</h3><p>Kırıntı madenciliği derya deniz bir konu. Hangi taşın hangi jeolojik koşullarda oluştuğunu, hangi eleme tekniğinin daha verimli olduğunu veya minerallerin yapılarını öğrenmek sürekli devam eden bir süreç.</p><p>Uygulamanın en değerli köşelerinden biri olan <strong>Akademi</strong> bölümü, tam olarak bu sürekli öğrenme ihtiyacını karşılamak için tasarlandı. Aşağıdaki ekran görüntüsünde de görebileceğiniz gibi, burada yer alan <strong>Bilgi Paylaşım Alanı</strong> sayesinde topluluğun her bir üyesi, kendi deneyimlerini ve araştırmalarını detaylı makaleler halinde yayınlayabiliyor.</p><ul><li><strong>Kendi Makalelerinizi Yazın:</strong> Ekrandaki sarı “Bilgi Paylaş” butonuna tıklayarak keşfettiğiniz bir mineralin özelliklerini, tarihçesini veya doğada nasıl bulunabileceğini anlatan kapsamlı yazılar kaleme alabilirsiniz.</li><li><strong>Bilgi Dağarcığınızı Genişletin:</strong> Diğer kullanıcılar tarafından hazırlanan “Ametist Nedir ve Nasıl Oluşur?” gibi içerikleri okuyabilirsiniz hatta bu tarz yazılar paylaşarak bilgilerinizi başka kullanıcılar ile de paylaşabilirsiniz.</li><li><strong>Kalıcı Bir Kaynak:</strong> Bu alan, akıp giden ve kaybolan anlık mesajlardan ziyade; Türkiye’nin ilk kırıntı madenciliği kütüphanesini kendi ellerimizle, hep birlikte inşa ettiğimiz kalıcı bir eğitim merkezi olarak işlev görüyor.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/688/1*pI1fYZcy0OSNG_raUFQucA.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü — 7</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/688/1*GseAHo7jKghlgPQG_xveBg.png" /><figcaption><a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura">Kırıntı Madencilik Topluluğu</a> Mobil Uygulamasından Ekran Görüntüsü — 8</figcaption></figure><h3>Serüvene Katılmaya Hazır Mısınız?</h3><p>Eğer doğada vakit geçirmeyi seviyorsanız, toprakta gizlenen tarihi ve mineralleri keşfetmek sizi heyecanlandırıyorsa, kırıntı madenciliğine başlamanın tam zamanı. Doğru araçlar ve destekleyici bir toplulukla, bir sonraki doğa yürüyüşünüz çok daha kazançlı ve keyifli bir maceraya dönüşebilir.</p><p>Türkiye’nin kırıntı madenciliği alanındaki bu ilk dijital adımının bir parçası olmak, keşiflerinizi paylaşmak ve yapay zeka asistanınızı hemen denemek için uygulamayı <a href="https://play.google.com/store/apps/details?id=com.enjoyz.goldenaura"><strong>Google Play Store’dan</strong></a> ücretsiz indirebilirsiniz.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0422cd385c77" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Silicon Valley’s Most Dangerous Fairy Tale: The ‘College Dropout’ Fetishism]]></title>
            <link>https://blog.startupstash.com/silicon-valleys-most-dangerous-fairy-tale-the-college-dropout-fetishism-cf4849ea1312?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/cf4849ea1312</guid>
            <category><![CDATA[mark-zuckerberg]]></category>
            <category><![CDATA[college-dropout]]></category>
            <category><![CDATA[bill-gates]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[steve-jobs]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Sun, 04 Jan 2026 12:34:11 GMT</pubDate>
            <atom:updated>2026-01-09T08:57:01.605Z</atom:updated>
            <content:encoded><![CDATA[<p>They didn’t succeed because they dropped out. They dropped out because they were already succeeding. Stop chasing a ghost story. It’s time to look at the data behind the legend.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*F_ll0NKfeQ9K5ZCYt9kaOg.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p><strong>Mark Zuckerberg, Bill Gates, Steve Jobs… When we hear these names, we don’t just think of trillion-dollar companies; we think of the legendary “dropout” narrative. But why does the system insistently whisper that degrees are obsolete? Is success really hidden in walking out the campus gates, or are we just victims of a massive statistical illusion?</strong></p><p>In the entrepreneurial world, especially within the tech ecosystem, a university degree has almost begun to be perceived as a symptom of lethargy. “Real visionaries don’t waste time in school; they build.” This has become the most popular motto of modern times.</p><p>However, as a founder, when you scratch the surface of this narrative, you don’t find a shiny formula for success. Instead, you uncover a massive logical fallacy known in scientific literature as <strong>“Survivorship Bias.”</strong></p><p>Let’s dissect this “College Dropout” myth under the light of data and cold, hard facts.</p><h3>The Scientific Backbone: Survivorship Bias</h3><p>During World War II, the military wanted to reinforce the armor of planes returning to base. They looked at where the returning planes had the most bullet holes and decided to armor those spots. The mathematician Abraham Wald stopped them. He pointed out that the data was flawed because they couldn’t see the planes that <em>didn’t</em> return. The planes hit in critical areas (engines, cockpit) had already crashed. The ones with bullet holes in the fuselage survived because those areas weren’t critical.</p><p>Entrepreneurship is exactly the same.</p><p>We only see the 0.001% of “survivors” who dropped out and became billionaires. We don’t see the tens of thousands of young people who dropped out and failed, fell into debt, or settled for mediocre careers. The media doesn’t like the headline <em>“Dropout Goes Bankrupt.”</em> But <em>“Dropout Becomes Billionaire”</em> always gets the click.</p><p>This “selective data” wires a false cause-and-effect relationship in our brains: <em>Dropping out = Success.</em> The real equation is never that simple.</p><h3>“Dropout”… But From Where? (Signaling Theory)</h3><p>The biggest trick of this myth is omitting the detail of <em>where</em> they dropped out from. Bill Gates and Mark Zuckerberg left <strong>Harvard</strong>. Evan Spiegel left <strong>Stanford</strong>.</p><p>These individuals had already passed through the world’s toughest academic filters. They were already in a pool of elite intelligence and networking. Their decision to leave wasn’t due to incompetence; it was because the <strong>Opportunity Cost</strong> of staying in school was higher than the value of the degree at that specific moment.</p><p>In economics, this is called <strong>“Signaling Theory.”</strong> Having been accepted into Harvard already sends a powerful enough signal to the market saying, “I am intelligent and disciplined.” The diploma itself had lost its marginal utility for these specific profiles. However, dropping out of an average university with average grades and no billion-dollar idea in hand does not send the same “signal.”</p><h3>What Do the Statistics Say?</h3><p>When we look at the data, the “College Dropout” legend has no statistical backing.</p><ul><li>According to research by the <strong>Kauffman Foundation</strong>, over <strong>90%</strong> of successful tech entrepreneurs are university graduates.</li><li><strong>84%</strong> of the billionaires on the Forbes 400 list hold a university degree.</li><li>More interestingly, in the age of <strong>“Deep Tech”</strong> (Biotech, Space, AI Hardware), a significant portion of founders hold PhD-level education.</li></ul><p>In 2026, building an OpenAI, a SpaceX, or a Moderna requires more than just coding in a garage; it requires deep academic accumulation and interdisciplinary engineering. Saying “I know how to code” isn’t the leverage it was in 2010 because AI can code now, too. The intellectual depth required to know <em>what</em> to build (which often comes with rigorous education) is now the scarcer asset.</p><h3>Why Does the System Glorify This?</h3><p>So, if the data is clear, why is this myth still so popular?</p><ol><li><strong>The Hero’s Journey:</strong> People love rule-breakers, not rule-followers. The young founder who “rebels” against the academic system is the modern-day Robin Hood.</li><li><strong>Cheap Labor &amp; Fast Cycles:</strong> Silicon Valley investors (think of the Peter Thiel Fellowship) love it when young people dedicate all their energy and time to their startups without the distractions of academic career anxiety. The education system prepares youth to be “corporate employees,” while VCs incentivize them to be “unicorn hunters.”</li><li><strong>Selling Hope:</strong> Not everyone can get into Harvard. Not everyone can be a genius. But everyone can “drop out.” This idea sells the illusion that success is accessible. Saying “school is useless” is a fantastic excuse for failure.</li></ol><h3>The Other Side of the Coin: The Educated Titans</h3><p>While the media talks about Zuckerberg, we forget the “degreed” geniuses running the world’s most valuable companies.</p><ul><li><strong>Jensen Huang (NVIDIA):</strong> Master’s degree from Stanford. Founded the most important chip company on earth.</li><li><strong>Satya Nadella (Microsoft):</strong> Master’s in Computer Science and an MBA. Resurrected Microsoft.</li><li><strong>Sundar Pichai (Google):</strong> Wharton Scholar.</li><li><strong>Tim Cook (Apple):</strong> Industrial Engineering and MBA.</li></ul><p>These names used the discipline, network, and analytical framework provided by education to manage empires.</p><h3>Conclusion: What Should You Do?</h3><p>There is no single path in entrepreneurship. If you have a product that the market demands insanely, and school is merely slowing you down, then a diploma is no different than a piece of paper. Drop out and build.</p><p>But dropping out just to be “cool,” without a Plan B, and without developing your intellectual capital, is not bravery it’s gambling.</p><p>Entrepreneurship is about taking risks, but <strong>“calculated”</strong> risks. Ignoring statistics, science, and your own reality to take a leap won’t make you the next Steve Jobs; it will likely push you to the invisible side of the statistics the crashed planes.</p><p>Remember: When Zuckerberg dropped out, Facebook had already taken over the campus. <strong>What do you have in your hand?</strong></p><p><em>This article is written by the founder of </em><a href="https://whowearblacks.substack.com/"><strong><em>Who Wear Blacks</em></strong></a><em>, offering a critical perspective on the world of entrepreneurship and technology.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cf4849ea1312" width="1" height="1" alt=""><hr><p><a href="https://blog.startupstash.com/silicon-valleys-most-dangerous-fairy-tale-the-college-dropout-fetishism-cf4849ea1312">Silicon Valley’s Most Dangerous Fairy Tale: The ‘College Dropout’ Fetishism</a> was originally published in <a href="https://blog.startupstash.com/">Startup Stash</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The End of Digital Privacy: X, Deepfake Pornography, and a New Generation of Violence Against Women]]></title>
            <link>https://medium.com/@enes.onion/the-end-of-digital-privacy-x-deepfake-pornography-and-a-new-generation-of-violence-against-women-dd0dc0769234?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/dd0dc0769234</guid>
            <category><![CDATA[tyranny]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[womens-rights]]></category>
            <category><![CDATA[grok]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Sat, 03 Jan 2026 14:51:39 GMT</pubDate>
            <atom:updated>2026-01-03T15:18:36.326Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dFsCYcpL0welvJMCfuJgiQ.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>For the past few months, when I grab my morning coffee and refresh my X (formerly Twitter) feed, instead of encountering exciting news about technological advancements, I am met with a dystopia that knots my stomach. Artificial Intelligence, once praised for its potential to cure cancer or facilitate our journey to Mars, has today transformed into the most terrifying weapon of cyberbullying in our arsenal.</p><p>Especially recently, the “Deepfake” epidemic, where the faces of women famous or not are mounted onto pornographic content without their consent, is spreading like an uncontrolled wildfire. In this article, I want to analyze how this technology has turned into a digital weapon, the deep cracks it creates in women’s psychology, and the inevitable end awaiting X if it fails to take radical precautions.</p><h3>Not “Just a Photo”: The Horrifying Picture Behind the Numbers</h3><p>Those who view this phenomenon merely as a “bad prank committed by a few individuals” are in a state of grave delusion. The statistics slap the severity of the situation right in our faces. According to data from 2023 and 2024, approximately <strong>98% of deepfake videos on the internet are pornographic content.</strong> Even more chilling is that <strong>99% of these targets are women.</strong></p><p>The Taylor Swift incident that recently caused a global uproar was actually just the tip of the iceberg. Fake images of Swift garnered 45 million views on X within just 17 hours. The platform was so slow to remove the images that their spread rate rivaled that of biological viruses.</p><p>However, the real danger lies not with celebrities, but with the voiceless. The incident at a high school in New Jersey, where male students used AI to generate and spread fake nude photos of their female classmates, demonstrates how widely this technology has spread to the grassroots level. You no longer need high-level coding skills or powerful computers to produce a deepfake; a simple app or a Telegram bot is enough to manufacture this violence.</p><h3>Psychological Destruction: Virtual Image, Real Trauma</h3><p>At this point, we need to pause and address the psychological dimension within a scientific framework. Most people might say, “It’s not real anyway, why is it being exaggerated?” However, the human brain cannot always distinguish between “reality” and “realistic simulation” when processing visual trauma.</p><p>In psychology, this condition is defined as <strong>“Non-Consensual Intimate Imagery (NCII)”</strong> victimization. Research reveals that women exposed to deepfake pornography exhibit symptoms of Post-Traumatic Stress Disorder (PTSD) similar to individuals who have suffered physical sexual assault.</p><ol><li><strong>Violation of Bodily Integrity:</strong> Women feel they have lost control over their own bodies. The use of their bodies as “public property” in the digital world creates a deep sense of helplessness and humiliation.</li><li><strong>Social Isolation and Paranoia:</strong> Victims withdraw from social life, delete their digital footprints, and give up on their careers out of fear: “Did my colleague see this image? Did my family see it?”</li><li><strong>Suicidal Tendencies:</strong> Especially in adolescent victims, this type of character assassination, combined with an unmanageable sense of shame, can trigger suicidal thoughts.</li></ol><p>This is not simple “trolling” behavior; this is a <strong>form of sexual violence</strong> where technology is used as leverage.</p><h3>X’s Silence and the “Wild West” Syndrome</h3><p>Let’s come to the platform dimension of the event. Elon Musk’s understanding of “free speech absolutism” has turned X into a “Wild West” town devoid of moderation. The massive layoffs within the Trust and Safety teams paved the way for such content to circulate unchecked.</p><p>Algorithms currently prioritize “engagement” above all else. The shockwave created by deepfake content drives retweets and likes. The algorithm doesn’t care whether the content is ethical; it looks at how long it stays on the screen. This creates a toxic cycle that rewards harassers.</p><h3>The Future Awaiting X: Inevitable Collapse?</h3><p>My personal opinion and strategic analysis is this: If X remains “reactive” (reacting after the event) rather than “proactive” (preventing the event), a dark future awaits the platform.</p><ol><li><strong>Advertiser Exodus:</strong> No major brand (Coca-Cola, Apple, etc.) wants their logo to appear right next to non-consensual pornographic content. Brand Safety is a red line. If X does not tighten moderation, it will turn into a “junkyard” where only gambling sites and crypto scammers advertise.</li><li><strong>Legal Squeeze and EU Sanctions:</strong> The European Union’s Digital Services Act (DSA) does not joke around. Platforms are obliged to remove illegal content rapidly. X’s nonchalance could lead to billions of dollars in fines and even access bans in the EU.</li><li><strong>“Dead Internet” and User Loss:</strong> A platform where women do not feel safe ceases to be mainstream. The exodus of female users and quality content creators will turn X into an echo chamber for marginal groups. This is the death warrant for a social media platform.</li></ol><h3>Conclusion: Technology Must Meet Ethics</h3><p>As someone within the software and tech world, I say this: AI is a great tool, yes. But every technological revolution that is not grounded in an ethical framework is a potential disaster.</p><p>What X needs to do is clear:</p><ul><li>Integrate algorithms that detect and watermark AI-generated images.</li><li>Show “zero tolerance” to accounts sharing “Non-Consensual Intimate Imagery” by banning them permanently (IP ban, not just shadowban).</li><li>Rebuild moderation teams with human oversight.</li></ul><p>As users, we must not just “scroll past” this digital violence; we must report it and socially isolate those who spread this content. Because what happens to a stranger today could happen to you, your spouse, or your daughter tomorrow. And when that day comes, we might painfully realize how insufficient it is to say, “it’s just an image.”</p><p>The internet is a mirror; it reflects whatever we give it. The reflection we currently see in that mirror gives us cause to be ashamed of our humanity. Changing this is, again, in our hands.</p><h3>What are your thoughts on this issue? Do X’s current policies make you feel safe? Let’s discuss in the comments, because remaining silent is consent.</h3><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dd0dc0769234" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Opening HR’s “Black Box”: How the EU AI Act is Reformatting the Recruitment World]]></title>
            <link>https://medium.com/@enes.onion/opening-hrs-black-box-how-the-eu-ai-act-is-reformatting-the-recruitment-world-d90142974b8b?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/d90142974b8b</guid>
            <category><![CDATA[human-resources]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[eu-ai-act]]></category>
            <category><![CDATA[future-of-work]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Thu, 01 Jan 2026 21:40:52 GMT</pubDate>
            <atom:updated>2026-01-01T21:40:52.784Z</atom:updated>
            <content:encoded><![CDATA[<p>As “Who Wear Blacks,” this week we suit up to go behind the scenes, straight into the courtroom of algorithms.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-5am_M4qnVijkAqOOQEHhQ.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>Imagine applying for a job. You upload your CV, hit the “Submit” button, and before you can even take a sip of your coffee, a cold notification pops up on your screen: <em>“Unfortunately, your qualifications do not match this position.”</em></p><p>Who rejected you? Who measured your skills, experience, and potential in those few split seconds? The answer, most likely, was not a human. For years, algorithms operating inside a “black box” where we don’t know how or why they choose what they choose have become the secret judges of our careers.</p><p>But the rules of the game are changing. The <strong>EU AI Act</strong>, recently adopted by the European Union, is saying “Stop, show your ID” to one of the most controversial areas of the business world: HR technologies. For employers, saying “I used AI” is no longer a show of innovation; it is now a serious legal responsibility and a test of transparency.</p><h3>The End of the Wild West: Why HR Was Declared a “High-Risk” Zone</h3><p>AI technologies have long developed in a regulatory vacuum a “Wild West” environment, so to speak. Companies integrated systems that scanned thousands of CVs in seconds for the sake of efficiency. However, when the EU Act classified technology based on usage areas, it placed HR and employment processes squarely in the <strong>“High-Risk”</strong> category.</p><p>Why? Because if a Spotify algorithm recommends the wrong song, it’s just annoying. But if an HR algorithm wrongly filters you out, it is a direct attack on your livelihood, your career, and your social status.</p><p>The legislator is saying this: <strong>“A human being’s future cannot be entrusted to an unchecked piece of code.”</strong></p><p>This classification imposes the following obligations on companies (employers) using recruitment tools:</p><ul><li>Risk management systems must be established.</li><li>Data governance must be flawless (Is the training data dirty or clean?).</li><li>Technical documentation and record-keeping are mandatory.</li><li>Transparency and human oversight are required.</li></ul><h3>The Transparency Paradox: “Was I Evaluated by a Robot?”</h3><p>The most radical change brought by the Act is the shift in power balance between the employer and the candidate. Until now, candidates were in the dark about how their applications were assessed. Now, <strong>Transparency Obligations</strong> come into play.</p><p>Employers must now be prepared for this scenario:</p><blockquote><strong><em>Candidate:</em></strong><em> “My application was rejected. Did a human make this decision, or was it AI?”</em></blockquote><blockquote><strong><em>Employer:</em></strong><em> “An initial screening was performed by an AI system.”</em></blockquote><blockquote><strong><em>Candidate:</em></strong><em> “Okay, on what parameters was I filtered out? What was the logic?”</em></blockquote><p>Here is the crux of the matter. The Act requires employers (and the providers of the systems they use) to make the logic of the decision <strong>“explainable.”</strong> If you are using a complex Deep Learning model and even its engineers cannot fully explain how it reached a decision, you cannot use that system.</p><p><strong>The “Black Box” era is ending; the “Glass Box” era is beginning.</strong> Candidates must be explicitly informed that they are interacting with AI. This means we will soon see disclaimers in job ads like: <em>“AI-assisted analysis tools are used for skill matching in the preliminary evaluation process of this position.”</em></p><h3>Algorithmic Bias: Coding Past Mistakes into the Future</h3><p>The fundamental fear lying beneath the “high risk” label mentioned at the start is <strong>Bias</strong>.</p><p>Let’s remember Amazon’s famous failed experiment. Years ago, Amazon developed an AI to scan CVs. However, the system’s training data (applications from the previous 10 years) consisted predominantly of men. The AI established a correlation that “successful employee = male” and penalized CVs containing the word “women’s” (e.g., “Women’s Chess Club Captain”).</p><p>The EU AI Act aims to prevent exactly this. Employers must prove that the systems they use:</p><ol><li>Have training data that is representative, error-free, and complete.</li><li>Do not discriminate based on gender, race, ethnicity, religion, or political opinion.</li><li>Are continuously tested.</li></ol><p>“The algorithm thought so” is no longer a defense; it could be a confession of guilt.</p><h3>Human-in-the-Loop: The Return of the Human</h3><p>Perhaps the most philosophical aspect of the Act is that it shakes the blind trust in technology. The law mandates <strong>“Human Oversight.”</strong></p><p>What does this mean? In a recruitment process, the final “reject” or “accept” decision cannot be left to a fully autonomous system. At a critical point in the process, a competent, trained human with the authority to override the system’s output must intervene.</p><p>This means employers cannot downsize HR departments to hand the job over entirely to robots. On the contrary, they will need more qualified HR specialists (perhaps “Algorithmic HR Auditors”) who can interpret the outputs of those robots. AI must remain a “decision support mechanism,” not the “decision-maker.”</p><h3>Emotion Recognition Ban: Your Facial Expressions Are Safe</h3><p>Another striking clause for workplaces is the ban on <strong>“Emotion Recognition”</strong> systems.</p><p>Some companies were attempting to measure “stress levels,” “honesty,” or “cultural fit” by analyzing candidates’ facial expressions, voice tones, and eye movements during video interviews. The EU Act views the use of these technologies in the workplace and education as an <strong>“Unacceptable Risk”</strong> and prohibits them.</p><p>Because a person being nervous in an interview does not mean they are incompetent. Or, someone from a different cultural background avoiding eye contact does not indicate dishonesty. AI’s pseudo-scientific deductions regarding these matters are no longer legal.</p><h3>The Global Domino Effect: Why Silicon Valley is Watching Brussels</h3><p>To say, “This law only binds companies in Europe, it doesn’t affect us,” is to misunderstand how the digital economy works. While this law is within EU borders, its impact will cross oceans. This is exactly the phenomenon known in legal and tech literature as <strong>“The Brussels Effect.”</strong></p><p>Remember GDPR (General Data Protection Regulation)? When it came into force, it didn’t just protect Europeans’ data. Giants like Google, Facebook, and Microsoft integrated these standards into their global operations because splitting their systems (safe for Europe, unsafe for the rest) was too costly and complex. Today, wherever you are in the world, your data is processed with protocols close to GDPR standards.</p><p>The same scenario is now repeating with the AI Act:</p><ul><li><strong>The Single Global Standard Necessity:</strong> Multinational Corporations (MNCs) cannot use a fragmented structure in recruitment processes. Training and maintaining a separate AI algorithm for the New York office and another for the Berlin office is an operational nightmare. Therefore, they will adopt the “strictest” standard set by the EU as the “gold standard” for all their global operations. This means a candidate in San Francisco will indirectly fall under the protection of this law.</li><li><strong>The Supply Chain Effect:</strong> The global HR tech market (Workday, SAP, Oracle, LinkedIn, etc.) is massive. None of these companies want to lose the wealthy market of 450 million people in the EU. So, they will make their algorithms “EU AI Act Compliant.” Once they make this update, they will sell this software to clients in Singapore or Canada with the same standards.</li><li><strong>Globalization of Legal Risks:</strong> It’s not just Europe; the US is waking up too. New York City’s <strong>“Local Law 144”</strong> mandates “bias audits” for automated tools used in hiring. When the world’s two major economic hubs (EU and New York) start speaking the same language, this becomes a universal business practice, not just a local rule.</li></ul><p>In short, the moment this law passed, it appeared on the desk of the developer in Silicon Valley and the HR Director in Asia. Because in the digital world, borders do not work as they do on maps.</p><h3>Conclusion: A New Employer Brand is Born</h3><p>Employers should not view this law as a shackle or a bureaucratic hurdle. On the contrary, this is an opportunity for <strong>trust building.</strong></p><p>The most desirable companies of the future will not be those “using the best AI,” but those <strong>“using AI in the most ethical, transparent, and human-centric way.”</strong> Being able to tell candidates, “You are being evaluated not by a pile of code, but by technology supported by human values,” will become the biggest competitive advantage.</p><p>For “<a href="https://whowearblacks.substack.com/">Who Wear Blacks</a>” readers, in summary: Technology is evolving, but the reins (and the responsibility) are passing back to human hands. Put on your black suits, but make that “black box” glass. Let the inside be seen.</p><h3>Note to the Reader:</h3><blockquote>Have you ever questioned whether the HR technologies used in your company are compliant with this law? Or are you still one of those saying, “The algorithm is always right”? Let’s discuss in the comments.</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d90142974b8b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[33 Million Reasons to Doubt: Was the River Ending Too Perfect? Coupang Case]]></title>
            <link>https://medium.com/@enes.onion/33-million-reasons-to-doubt-was-the-river-ending-too-perfect-coupang-case-dad8f0978254?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/dad8f0978254</guid>
            <category><![CDATA[technews]]></category>
            <category><![CDATA[coupang]]></category>
            <category><![CDATA[data-leak]]></category>
            <category><![CDATA[korea]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Mon, 29 Dec 2025 00:17:50 GMT</pubDate>
            <atom:updated>2025-12-29T00:17:50.544Z</atom:updated>
            <content:encoded><![CDATA[<p>When a data breach story sounds like a spy movie script, we have to ask: Are we watching the resolution, or just a distraction?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_1z8U3nJtXNPN4F46gRnPw.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>A man stands with his head bowed before the cameras: Kim Bom-suk, the founder of South Korean e-commerce giant Coupang. A delayed apology, vague promises of compensation, and a sentence spilling from his lips that rivals the absurdity of a cheap spy novel: <em>“The data didn’t leak because the laptop got wet in the river.”</em></p><p>Most people are focused on the word “compensation” in the headlines, calculating the 100,000 Won (approx. $70 USD) they might receive. But as an entrepreneur and an observer watching the dark back alleys of technology, I am telling you: what is <em>not</em> being discussed in this case is far more terrifying than what is.</p><p>I am not a victim of this breach, nor am I a South Korean citizen. But the picture I see from the outside is a global warning flare that concerns not just Korea, but anyone with a digital footprint.</p><p>Let’s take a closer look at the “official theater” on the table, and the dark “intelligence analysis” hiding right behind it.</p><h3>The Play on Stage and Inconsistent Numbers</h3><p>Kim Bom-suk’s apology came with strategic timing, right when parliamentary pressure and public outrage were hitting their peak. They stated a compensation plan would be announced “as soon as possible.” Yet, the details remain foggy. While victims expect cash compensation per person, rumors suggest the company will try to gloss over the incident with “discount coupons” and “site points” to cut costs.</p><p>However, the real devil is hiding in the details of the numbers.</p><p>Coupang admitted that the attacker (a former employee) “accessed” the data of <strong>33.7 million</strong> customer accounts. This equates to almost the entire adult population of South Korea. Names, phone numbers, addresses…</p><p>But the company offers a defense that insults the intelligence: <em>“Yes, 33 million records were accessed, but the attacker only saved 3,000 of them to his personal computer.”</em></p><p>Stop and think for a moment. You break into a bank vault, gain access to a room filled with billions of dollars, but you only take three coins and leave? This goes against the nature of cybercrime, the psychology of theft, and the mathematics of the data trade. Usually, such data is backed up to cloud servers in seconds.</p><h3>Wet Hard Drives and the Illusion of “Proof”</h3><p>Here is where the story gets its most “Hollywood” treatment. Coupang claims the following:</p><ol><li>The suspect who stole the data (reported to be a Chinese national) threw his laptop into a river when he realized he would be caught.</li><li>Coupang hired “special divers” to retrieve this laptop from the water.</li><li>From that soaked, likely short-circuited device, they determined that “the data was not sent to anyone else.”</li></ol><p><strong>Analysis:</strong> There are massive cracks in this scenario that prevent us from being “truly sure.”</p><ul><li><strong>Forensic Impossibility:</strong> Cleaning, drying, recovering, and analyzing data from a device pulled from water is a delicate laboratory process that takes weeks. For Coupang to claim they verified this “immediately” is technically suspicious.</li><li><strong>Government Denial:</strong> The moment Coupang claimed, “We cooperated with the government,” the South Korean Ministry of Science and ICT, along with the Police, issued a harsh correction: <em>“This is not a result we confirmed; it is Coupang’s own claim.”</em></li><li><strong>The Suspect’s Location:</strong> If the attacker is not currently in South Korean police custody (and there is a high probability he has fled abroad), his declaration that “I didn’t sell the data” holds absolutely no value.</li></ul><h3>The “Silent Phase” of the Digital Apocalypse</h3><p>So, let’s assume Coupang is wrong (or lying) and this data actually leaked. But we don’t see this data on Dark Web forums or hacker sites. Is this good news?</p><p>No. This is the most dangerous phase of chaos.</p><p>The period when data is not “dumped” publicly is the period when it is being <strong>processed, categorized, and sold in “closed loops” to the highest bidder.</strong></p><p>If the data of millions of people is currently sitting on unknown third-party servers, here are the scenarios awaiting us:</p><h4>A. “Whaling” and Wealth Maps</h4><p>If data is public, its value is $1. But if it is held by a select group, it is a weapon. Criminal organizations may currently be creating “Wealth Maps” with this data. They analyze who lives in Seoul’s most luxurious districts, and how often they order expensive watches or electronics. This goes beyond a simple cyber attack; it is an intelligence operation for <strong>physical extortion, kidnapping, and blackmail.</strong></p><h4>B. AI-Powered Social Engineering</h4><p>Scam emails used to start with “I am a Prince,” and we would laugh. Now, data meets AI. Using leaked phone numbers and customer service voice recordings, we are facing algorithms that can mimic victims’ voices (<strong>Deepfake Audio</strong>) and know their most intimate spending habits. A world where a business owner’s accountant gets a call from his “boss” ordering an urgent wire transfer is no longer science fiction.</p><h4>C. “Zombie” Identities</h4><p>Professional attackers don’t change passwords immediately. They watch. They wait for the victim’s most vulnerable moment — when they are on vacation, in the hospital, or have limited internet access. While the victims continue with their lives, their “digital twins” could be establishing shell companies, laundering money, or taking out loans in the background.</p><h3>Conclusion: “Zero Trust” Is No Longer a Choice</h3><p>The Coupang case shows us that the “wet laptop pulled from the river” might just be a decoy. Until official authorities explicitly say “it is clean,” we must assume the data has been stolen and is currently being processed.</p><p>This analysis screams one thing to the world: The question “Will my data be stolen?” is obsolete. The real question is: <strong>“My data was stolen long ago; who is processing it right now, and when will they come knocking?”</strong></p><p>This situation forces us all into a <strong>“Zero Trust”</strong> societal structure where we trust no one and no system. In this new world, being paranoid is not a disorder; it is a survival reflex.</p><p>Remember: In this new world, being paranoid is not a disorder; it’s a survival reflex.</p><p><strong>Enes</strong> <em>Founder, </em><a href="https://whowearblacks.substack.com/"><em>Who Wear Blacks</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dad8f0978254" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Anatomy of a Fall: The $600M Unicorn Autopsy]]></title>
            <link>https://blog.startupstash.com/anatomy-of-a-fall-the-600m-unicorn-autopsy-9a6bb55806f1?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/9a6bb55806f1</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[fall]]></category>
            <category><![CDATA[ÿnsect]]></category>
            <category><![CDATA[unicorns]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Sat, 27 Dec 2025 20:31:01 GMT</pubDate>
            <atom:updated>2025-12-29T06:43:41.854Z</atom:updated>
            <content:encoded><![CDATA[<p>The cost of treating biology like software: The Ÿnsect case and the hard truths of hardware entrepreneurship.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BlF-gRQIQ2SvRZ9wde-T-g.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>It was the “golden child” of the European startup ecosystem. Backed by the vision of “Iron Man” Robert Downey Jr., the endless support of the French government, and over $625 million in cash.</p><p><strong>Ÿnsect</strong> wasn’t just farming insects; it promised a revolution to solve the food crisis by combining vertical farming, robotics, AI, and biology. However, news late in 2024 revealed that this revolution had hit a wall. The company filed for a “safeguard procedure” in France, laid off staff en masse, and was forced to radically change its strategy.</p><p>So, how did this project, which looked flawless on paper and had VCs sleeping at its doorstep, crash into the cold ground of reality? This isn’t just the failure of one company; it is a tragic story of “Treating Biology Like Code” and “Premature Scaling.”</p><p>Let’s break down this $600 million case study for entrepreneurs.</p><h3>The Big Promise: Being the “Tesla of Agriculture”</h3><p>Ÿnsect’s business model was based on a very logical foundation: The world population is growing, the need for protein is exploding, but current livestock farming is killing the planet. The solution? <strong>Mealworms.</strong></p><p>Mealworms are incredible; they consume little water, can live in millions stacked on vertical shelves, and turn into high-quality protein. Ÿnsect’s plan was to build the world’s largest vertical insect farm in Amiens. A 40-meter-high “Giga-Factory,” managed entirely by autonomous robots, producing 200,000 tons of insects a year.</p><p>Investors loved it. Because this wasn’t just agriculture; it was a tech play. But there were massive details that were overlooked.</p><h3>Where Did Reality Hit? (Technical and Financial Details)</h3><p>Ÿnsect’s fall cannot be attributed to a single reason. It was a “Perfect Storm” scenario.</p><p><strong>A. Biology is Not Software (The Scale-Up Trap)</strong> The Silicon Valley logic is: “Write code once, if it works, copy it to 1 million users.” The cost is near zero. But biology doesn’t work that way.</p><ul><li><strong>Heat and Disease:</strong> It’s easy to raise insects in a lab (in small containers). But when you try to do this with millions of insects in a massive 40-meter-high hangar, ambient temperature, humidity balance, and airflow become chaotic.</li><li><strong>Result:</strong> Ÿnsect faced unexpected biological resistance scaling living organisms at the Amiens factory. The growth rates and survival rates of the insects did not match the Excel spreadsheets.</li></ul><p><strong>B. CAPEX Hell (Capital Expenditures)</strong> Instead of developing the technology, Ÿnsect chose to <strong>build the factory with its own money.</strong> Construction costs skyrocketed with global inflation and supply chain crises. Initial budgets were exceeded. Most of the $600 million was buried in steel structures and concrete, not R&amp;D.</p><p><strong>C. Energy Crisis and Macroeconomics</strong> Vertical farming is “energy-intensive.” You need to keep the environment at a specific temperature constantly. when energy prices in Europe went through the roof after the Russia-Ukraine war, Ÿnsect’s Unit Economics became unsustainable. Competitors (traditional soy or fishmeal) remained cheaper.</p><h3>Critical Lessons for Entrepreneurs</h3><p>There are very clear lessons to be learned from this story. If you are in hardware, agriculture, or manufacturing, pin these to your wall.</p><ul><li><strong>Lesson #1: The Gap Between TRL (Technology Readiness Level) and Scaling</strong> Ÿnsect invested in TRL 9 (Full Commercial Production) before its technology was fully mature (at TRL 7–8). Never start “Full Scale” when building “First of a kind” facilities. If your pilot plant isn’t profitable, your main factory never will be. Don’t scale your losses.</li><li><strong>Lesson #2: Product-Market Fit and Pricing</strong> Ÿnsect’s main product was animal feed and fertilizer (Commodity). But its cost structure was high-tech, robotics, and expensive engineers (Ferrari). The customer doesn’t care how technological your insect is; they look at the price and protein ratio. If you are doing High-Tech manufacturing, you must either drive costs down incredibly or sell to a “Premium” market.</li><li><strong>Lesson #3: Don’t Be Late to Pivot the Business Model</strong> When cash flow started drying up, Ÿnsect tried to change its business model. They said, “Let’s not be farmers, let’s license this technology to others.” However, you cannot license technology to someone else before proving it works flawlessly in your own factory.</li></ul><h3>Conclusion: Is DeepTech Dead? (Website and Vision Analysis)</h3><p>Looking at the Ÿnsect case and saying “DeepTech was a bubble, it burst” is taking the easy way out. The reality is this: <strong>Investors no longer want to invest in concrete, steel, and bricks.</strong> DeepTech isn’t dying, it’s just shedding its skin.</p><p>When we examine the company’s website and new brands (<strong>Ynfabre</strong> and <strong>Sprÿng</strong>), we see that Ÿnsect has realized this lesson, albeit late. The company is giving up the claim of being “The world’s largest insect farmer” and trying to become <strong>“The Intel of the insect world.”</strong></p><ul><li><strong>Old Model:</strong> Find money, build factory, produce insects, sell. (High Risk, High Capital)</li><li><strong>New Model:</strong> Develop technology and genetics (<strong>Ynfabre</strong>), create a “Premium” brand (<strong>Sprÿng</strong>), let local partners or licensees handle production. (Low Capital, High Tech)</li></ul><p><strong>The Golden Rule for Entrepreneurs:</strong></p><p>If you are founding a DeepTech startup, ask yourself this question: <strong>“Am I digging for gold, or am I selling shovels?”</strong></p><p>Ÿnsect was a giant whose pickaxe broke while trying to dig for gold. Now, it is trying to survive by selling its most valuable asset: the technology to make “stronger pickaxes” (IP and Genetics).</p><p>Just as the “SaaS” (Software as a Service) model rules the software world, the future for the physical world lies in the <strong>“TaaS” (Technology as a Service)</strong> model. You don’t have to operate the factory; designing the factory’s brain is enough.</p><p>As a DeepTech entrepreneur, your job is to turn science into a commercial product, not to do civil engineering.</p><blockquote><em>Additional tools and resource ➡️ </em><strong><em>Visit </em></strong><a href="http://www.startupstash.com/"><strong><em>StartupStash</em></strong></a><strong><em><br></em></strong><em>Zendesk is giving $75,000 in credits and perks for startups! ➡️ </em><a href="https://tinyurl.com/4ta2c8j6"><strong><em>Apply Now!</em></strong></a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9a6bb55806f1" width="1" height="1" alt=""><hr><p><a href="https://blog.startupstash.com/anatomy-of-a-fall-the-600m-unicorn-autopsy-9a6bb55806f1">Anatomy of a Fall: The $600M Unicorn Autopsy</a> was originally published in <a href="https://blog.startupstash.com/">Startup Stash</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Deleted Files Never Die: NYPD, “Dirty Data” and the Dark Side of AI]]></title>
            <link>https://medium.com/@enes.onion/deleted-files-never-die-nypd-dirty-data-and-the-dark-side-of-ai-362bb679f24c?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/362bb679f24c</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[new-york-city]]></category>
            <category><![CDATA[intelligence]]></category>
            <category><![CDATA[muslim]]></category>
            <category><![CDATA[nypd]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Wed, 24 Dec 2025 01:20:36 GMT</pubDate>
            <atom:updated>2025-12-24T01:20:36.145Z</atom:updated>
            <content:encoded><![CDATA[<p>Do the ghosts of the “Muslim Surveillance Unit,” shut down 10 years ago, live on today in predictive policing algorithms? How does a child’s schoolyard fight or a family’s weekly grocery run get twisted by AI into “suspicion of terror”?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BM2ZThgudlRbXVAUPnm4IA.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>As I grabbed my morning coffee and scanned the global tech feed, a lawsuit headline stopped me in my tracks. The New York Police Department (NYPD) is back in court over data they claimed to have “deleted and destroyed” years ago. The issue isn’t just old files; it’s the impossibility of the “right to be forgotten” in the digital age, and worse, the likelihood that this <strong>“dirty data”</strong> is now feeding today’s powerful artificial intelligence systems.</p><p>Those who know me are aware of how obsessed I am with data sovereignty and digital privacy. Today’s news is not a dystopian movie scenario; it is the dirtiest truth of data management.</p><p>Let’s dive deep into this event and analyze why it concerns not just Muslims in New York, but everyone who generates data — especially those who mistakenly believe, “I have nothing to hide.”</p><h3>The Backstory: “Didn’t We Close That Book?”</h3><p>Let’s refresh our memories. After 9/11, the NYPD, with help from the CIA, established a secret unit: the <strong>Demographics Unit.</strong> This unit had a single mission: to map and track Muslim communities in and around New York.</p><p>In 2014, following immense pressure and lawsuits, this unit was disbanded. The NYPD promised to destroy the data they held. However, this new lawsuit filed in the final days of 2025 claims: <strong>“The NYPD lied. The data wasn’t deleted. In fact, this data forms the foundation of the next-generation surveillance technologies currently in use.”</strong></p><p>But do you think this database only contained “who went to which mosque”? You are mistaken. What is truly terrifying is how the most ordinary flow of life has been criminalized.</p><h3>Deep Analysis: The Tagging of Innocence (What Data Was Tracked?)</h3><p>If you think like an intelligence analyst, you aren’t looking for “crimes,” you are looking for “patterns.” The data the NYPD collected — and which is likely feeding AI right now — wasn’t criminal profiles, but the life cycles of ordinary families.</p><p>It’s not hard to guess what was stored and correlated in this “shadow database”:</p><h4>1. “Potential Threats” in School Corridors</h4><p>A minor scuffle involving a high school student is recorded as “teenage angst” for an ordinary student. However, for a Muslim youth in the target demographic, this data point is entered into the database tagged as <strong>“prone to violence”</strong> and <strong>“open to radicalization.”</strong> A teacher’s note saying “withdrawn/introverted” can be interpreted by the system as “isolating from society (suspicion of cell formation).” A political critique written in an essay by a 15-year-old turns into a “risk score” that prevents them from finding a job when they turn 25.</p><h4>2. Political Analysis of Grocery Receipts</h4><p>A family’s weekly shopping list is the most valuable intelligence source for an AI.</p><ul><li><strong>Why are they using excess cash?</strong> (Trying to leave no trace?)</li><li><strong>Why did they buy camping gear?</strong> (Weekend hiking, or paramilitary training preparation?)</li><li><strong>Why are they buying food in bulk?</strong> (A large group of guests, or an organizational meeting?) An ordinary “discount market run” can be coded as a “preparation phase” in the wrong algorithmic context.</li></ul><h4>3. “Leisure” Activities</h4><p>Gyms visited by young people, especially martial arts classes, are monitored under the suspicion of “physical conditioning.” Even family picnics in the park on weekends are mapped to determine “who talks to whom” and “which families come together.”</p><h3>The Big Picture: The Mosaic Effect and Algorithmic Bias</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fg8-BVAxzXucXCJB2eq2EA.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>Now, let me explain the magnitude of the danger with this formula:</p><blockquote><strong><em>(School Disciplinary Record) + (Dad’s Camping Tent) + (Uncle’s Mosque Association Membership) = High-Risk Terror Suspect</em></strong></blockquote><p>On their own, these data points are completely innocent. But when combined in an <strong>Artificial Intelligence (AI)</strong> pool, they turn into a cage of guilt from which escape is impossible. In intelligence, this is called the <strong>“Mosaic Effect.”</strong> Small, meaningless stones come together to form a large (and often incorrect) picture.</p><p>If the NYPD did not delete this “biased data” collected illegally between 2002 and 2014, and instead fed it as training data to a 2025 model <strong>Predictive Policing AI</strong>, that algorithm is saying today: <em>“According to my historical data, Muslim families buying camping gear should be monitored 85% more closely.”</em></p><p>Thus, human-made racism hides between lines of code, disguised as an unquestionable “mathematical reality.”</p><h3>Future Projection: The “Invisible Prison”</h3><p>Heading into 2026, this lawsuit shows us three critical things:</p><ol><li><strong>There is No Digital Eraser:</strong> Institutions do not delete data; they only archive it. Believing that data is truly gone is the greatest naivety of the digital age.</li><li><strong>AI Courts:</strong> In the future, courts will focus less on the crime itself and more on “what data trained the AI” that identified the suspect.</li><li><strong>Community Power:</strong> As an individual, it is impossible to stand against these massive surveillance mechanisms. In a world where even your grocery receipt can be evidence against you, the value of structures like <strong>Who Wear Blacks</strong> — communities that draw their own digital boundaries, advocate for data sovereignty, and protect their “private” spaces — will be vital.</li></ol><h3>Conclusion</h3><p>The NYPD case is not just a local New York issue. It is a giant warning sign for those of us living in, London,Berlin…</p><p>Your child’s grade at school, the fertilizer you bought from the supermarket, the camping trip you took on the weekend… It’s all accumulating somewhere. And tomorrow, when someone says, “Oops, we forgot to delete that data,” that data will have already become the building block of an AI that labels you a “potential criminal.”</p><p>Data security is not a technical detail; it is the safety of your family, your past, and your future.</p><p>Keep your eyes open, your data encrypted, and your community strong.</p><p><strong>– Enes</strong> <em>Founder, </em><a href="https://whowearblacks.substack.com/"><em>Who Wear Blacks</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=362bb679f24c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What the Spotify Leak Tells Us About the Future]]></title>
            <link>https://medium.com/@enes.onion/what-the-spotify-leak-tells-us-about-the-future-1bfc8c5d662e?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/1bfc8c5d662e</guid>
            <category><![CDATA[news]]></category>
            <category><![CDATA[hacking]]></category>
            <category><![CDATA[data-leak]]></category>
            <category><![CDATA[spotify]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Mon, 22 Dec 2025 20:53:02 GMT</pubDate>
            <atom:updated>2025-12-22T20:53:02.030Z</atom:updated>
            <content:encoded><![CDATA[<p>The biggest “library heist” in internet history has occurred. 86 million songs have dissolved into the dark waters of decentralized networks. But did this event just kill copyright, or did it kill creativity itself?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fcom-Ee2fLIhC4s3Er0Esw.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>December 2025 is a candidate to go down in the history books of the digital age as “The Great Data Rupture.” You have likely seen the headlines: <em>“Spotify archive leaked.”</em> However, reading this sentence and scrolling past is like looking only at the tip of an iceberg.</p><p>A group of activists, positioning themselves as “radical guardians of knowledge” and operating in close contact with <strong>Anna’s Archive</strong>, have breached Spotify’s digital ramparts. The result? Approximately <strong>86 million audio files</strong>, metadata for 256 million tracks, and a mind-boggling total data pile of <strong>300 Terabytes</strong>.</p><p>This isn’t the romanticized “sharing a few MP3s” of the Napster era. This is the auditory memory of human history being torn away from a corporate structure and hurled into uncontrolled chaos. Let’s dissect the technical anatomy, the economic aftershocks, and the dystopian AI scenarios awaiting us.</p><h3>Technical Anatomy: Not a “Hack,” but Systematic Exploitation</h3><p>The term “Hacked” used by the media oversimplifies the technical intelligence of the event. The attackers didn’t crack Spotify’s database passwords (Brute Force) or infiltrate their servers (Intrusion). What they executed was a much more sophisticated <strong>“Mass Scraping”</strong> operation.</p><ul><li><strong>How did the process work?</strong> The attackers likely used tens of thousands of “zombie” accounts and constantly rotating IP addresses (Residential Proxy Networks) to send requests to Spotify APIs that mimicked human behavior. By the time the system realized it was facing a bot army, it was too late.</li><li><strong>File Format Detail:</strong> Most of the leaked files are available in the <strong>Ogg Vorbis</strong> or <strong>AAC</strong> formats used by Spotify, in their raw forms with the encrypted DRM layers stripped away.</li><li><strong>The Metadata Treasure:</strong> More dangerous than the audio files is the <strong>Metadata</strong>. The BPM (beats per minute) of songs, genre tags, release years, popularity scores… This is the “big data” of the music industry, and it is now open for anyone to access.</li></ul><h3>The Ideological Dimension: The Ghost of Aaron Swartz and “Guerilla Open Access”</h3><p>To understand this act, one must look at the group’s manifesto. This group views digital content not as “commercial goods,” but as “cultural heritage.”</p><p>As internet activist Aaron Swartz wrote years ago in the <em>Guerilla Open Access Manifesto</em>: <em>“Information is power. But like all power, there are those who want to keep it for themselves.”</em></p><p>The attackers argue that Spotify (or Apple Music) could pull the plug one day, or that albums could be deleted due to licensing disputes (which happens often). To them, this 300 TB archive is a Noah’s Ark for music. However, is this act of “preservation” actually a “forced expropriation” that disregards the artist’s labor? This is exactly where the ethical gray area lies.</p><h3>AI and “Data Laundering” (The Darkest Scenario)</h3><p>The darkest scenario of this leak isn’t the end-user listening to free music; it is the <strong>training of Artificial Intelligence models.</strong></p><p>Currently, companies like Google (MusicLM), Suno, or Udio keep their datasets secret or work with limited licenses due to fear of copyright lawsuits. However, rules do not apply to the “Underground AI World” (Dark AI).</p><ul><li><strong>Model Training:</strong> These 86 million songs are the perfect dataset to teach an AI model the “mathematics of music.”</li><li><strong>Style Cloning:</strong> Using this data, AI can mathematically copy not just the voice, but the <em>composition style, chord progressions, and melodic preferences</em> of a specific artist (e.g., Taylor Swift or Daft Punk).</li><li><strong>The Industry Eating Its Own Tail:</strong> AI models trained with this data can generate thousands of new songs per second. When these songs are uploaded back to Spotify royalty-free, the platform will be flooded with bot music trained on its own data. Real human-produced music risks drowning in this ocean of synthetic noise.</li></ul><h3>Economic Analysis: The “Comet” Effect</h3><p>Spotify won’t go bankrupt financially from this event. User experience (UX), playlists, and social features are still more attractive than piracy. But who will take the real hit?</p><ul><li><strong>Independent Artists (Indie):</strong> Artists already earning $0.003 per stream will have to compete with AI-generated “Muzak” (elevator/background music). The leaked data will accelerate AI’s total takeover of this “background music” market.</li><li><strong>Data Value:</strong> Paradoxically, the value of clean, licensed data will increase. Companies will have to pay larger sums to obtain certification proving, “My model was not trained with this stolen data.”</li></ul><h3>The Legal Dead End: Suing a Decentralized Ghost</h3><p>Spotify and record labels (Universal, Sony, Warner) are currently deploying their most aggressive lawyers. But there is no company to face them.</p><p>The data is being distributed via the <strong>BitTorrent</strong> protocol and <strong>IPFS</strong> (InterPlanetary File System). There is no central server. The data is fragmented across thousands of users’ computers. If they shut down one “seed,” another will open. It’s like cutting off the head of a Hydra; two more grow back. The legal system is trying to fight 21st-century decentralized technology with 20th-century rules, and it is losing.</p><h3>Conclusion: Pandora’s Box Will Not Close</h3><p>This 300 Terabyte data leak should not be viewed merely as a “piracy event.” This is the <strong>final stage in the commodification of human creativity.</strong></p><p>Music has ceased to be a work torn from an artist’s soul and has become “fuel” (a dataset) that AI models need to consume. Anna’s Archive has spilled this fuel onto the streets. Now, everyone has a match.</p><p>The days are very near when we won’t be able to distinguish whether the heartbreak behind a new song belongs to a human or an algorithm fed by this leak. And perhaps the scariest part is that the majority won’t even care.</p><h4>Sources</h4><blockquote><a href="https://whowearblacks.substack.com/">whowearblacks</a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1bfc8c5d662e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Weekend Intelligence Report: December 20–21]]></title>
            <link>https://medium.com/@enes.onion/weekend-intelligence-report-december-20-21-c5480a30aba2?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/c5480a30aba2</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[intelligence]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[energy]]></category>
            <category><![CDATA[weekend]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Mon, 22 Dec 2025 12:46:31 GMT</pubDate>
            <atom:updated>2025-12-22T12:46:31.681Z</atom:updated>
            <content:encoded><![CDATA[<p>The Industrialization of AI: Energy Constraints, Capital Wars, and the Rise of Sovereign Tech…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hhcFvdqcdcqSFt9qVtnjcg.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><h3>The Intersection of Energy and AI: Fuel for the New Industrial Revolution</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hlqRras-mzwLn6itWbk6Jg.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>The defining theme of the final weekend of 2025 has been the formation of massive and unexpected alliances aimed at overcoming the energy bottleneck in artificial intelligence infrastructure. Technology companies are no longer just coding entities; they are transforming into “energy holdings” that build or finance their own power plants.</p><h4>1.1. Trump Media and TAE Technologies: The Triangle of Media, Politics, and Fusion</h4><p><strong>Anatomy of the Deal:</strong> The most sensational development of the week is the <strong>$6 billion</strong> merger agreement between <strong>Trump Media &amp; Technology Group (TMTG)</strong>, owned by President Donald Trump, and nuclear fusion developer <strong>TAE Technologies</strong>. Structured as an all-stock transaction, the deal stipulates that TMTG shareholders and TAE shareholders will each hold approximately 50% of the new entity.</p><p><strong>Strategic Logic and “Unusual Pairing”:</strong> At first glance, a merger between a social media platform (Truth Social) and a nuclear physics laboratory may seem illogical. However, upon deeper inspection, this move is a speculative arbitrage attempt targeting “energy,” the greatest constraint of the AI era.</p><ul><li><strong>The Energy Supply Problem:</strong> The electricity consumption of AI data centers is straining grids and limiting the growth of tech giants. TAE Technologies aims to produce “aneutronic” (neutron-free) fusion energy using hydrogen-boron fuel. This represents a theoretically limitless and clean energy source for AI data centers.</li><li><strong>Access to Capital Markets:</strong> Although TAE has raised $1.3 billion in private equity from giants like Google and Goldman Sachs, “deep tech” projects like fusion require constant cash flow. As a public company (NASDAQ: DJT), TMTG is using its market cap — inflated by “meme stock” dynamics — as leverage to fund TAE’s R&amp;D costs. Under the agreement, TMTG has committed to injecting $200 million in cash into TAE at signing.</li></ul><p><strong>Market Reaction and Future Projection:</strong> Following the news, TMTG shares gained over 30% in pre-market trading. This reaction indicates that investors are beginning to price Trump Media not just as a social media company, but as a potential candidate for an “energy giant.” However, the risks are immense; the projected timeline for fusion commercialization (2030s) presents a temporal mismatch with AI’s current energy hunger. If successful, this structure would become the world’s first publicly traded nuclear fusion company.</p><h4>1.2. Last Energy and the Modular Nuclear Future</h4><p>In the European and UK markets, a more pragmatic and immediately applicable solution is emerging over massive fusion reactors: <strong>Small Modular Reactors (SMRs).</strong></p><p><strong>Investment Details:</strong> Nuclear startup <strong>Last Energy</strong> has raised <strong>$100 million</strong> (Series B) to commercialize its micro-reactor technology developed to power AI data centers. Unlike traditional nuclear power plants, the company designs 20-megawatt (MWe) units that can be mass-produced in a factory and assembled on-site.</p><p><strong>Technological Differentiation and Market Strategy:</strong> Last Energy’s approach transforms a “civil engineering” problem into a “manufacturing engineering” problem. The reactors have the capacity to power 15,000 homes and offer <strong>off-grid</strong> capability by being installed directly next to data centers.</p><ul><li><strong>Risk Management:</strong> The company aims to minimize safety concerns by encasing the reactor core in a 1,000-ton steel shield and using a “sealed” design that requires no refueling during its operational life.</li><li><strong>Pilot Projects:</strong> While a 5 MW pilot reactor is planned in Texas, the primary commercial target is stated to be energy-intensive data centers in Europe. This is a solution that directly addresses Europe’s energy crisis and decarbonization goals.</li></ul><h4>1.3. The Battery Strategy of Asian Giants: Alibaba, Tencent, and ByteDance</h4><p>Chinese tech giants are focusing on renewable energy integration and battery storage technologies rather than nuclear energy. This is directly related to the Chinese government’s “Green Data Centers” mandate.</p><p><strong>Integration Strategy:</strong> Alibaba, Tencent, and ByteDance (TikTok’s parent company) are purchasing massive Battery Energy Storage Systems (BESS) to manage the energy consumption of their data centers.</p><ul><li><strong>Alibaba:</strong> As part of a $53 billion cloud and AI infrastructure investment plan, it is installing integrated “solar + storage” systems in its data centers. The company views storage as mandatory to manage the <strong>intermittency</strong> of renewable energy.</li><li><strong>Tencent:</strong> By establishing microgrids combining wind, solar, and energy storage at its Huailai data center, it has succeeded in meeting 71% of the facility’s energy needs through on-site generation.</li><li><strong>ByteDance:</strong> Stating that it needs 450–800 MWh of storage capacity for every 100 MW of new AI compute capacity, it has become a major buyer in the lithium-ion battery supply chain.</li></ul><p><strong>Sectoral Inference:</strong> These three examples (Fusion, SMR, Battery) prove that the global technology sector has now become a component of the “energy sector.” The message for investors is clear: 2026 portfolios must include the energy technologies feeding AI as much as the AI stocks themselves.</p><h4><a href="https://whowearblacks.substack.com/">WhoWearBlacks’ Strategic Takeaways:</a></h4><p>Developments in the final quarter of 2025 demonstrate that the “software eats the world” paradigm has ended, replaced by the reality that <strong>“energy constraints software.”</strong> The Trump Media–TAE merger, Last Energy’s SMR investments, and Asian battery moves all point to a single common denominator: <strong>Compute is no longer just a chip issue; it is an energy supply chain issue.</strong></p><p>This section of the report outlines strategic foresight and action plans for the entrepreneurial ecosystem based on these developments.</p><h4>1. New Market Dynamic: The “Energy-Compute” Asset Class</h4><p>The most distinct strategic outcome of the next 5 years is the inseparable coupling of energy generation and data processing capacity.</p><ul><li><strong>The New Definition of Vertical Integration:</strong> In the past, vertical integration was software meeting hardware (e.g., Apple). In the future, vertical integration will mean a data center owning its own nuclear reactor or microgrid.</li><li><strong>The Off-Grid Trend:</strong> The Last Energy example shows that entrepreneurs will focus on generating their own local energy solutions rather than waiting for cumbersome national power grids. Off-grid is not an operational preference but a necessity for sustainability and speed.</li></ul><h4>2. Which Entrepreneurs Does This Concern?</h4><p>This new conjuncture will directly affect not only energy giants but also the following three main entrepreneurial groups:</p><p><strong>A. Deep Tech and Cleantech Entrepreneurs</strong></p><ul><li><em>Opportunity:</em> Generating energy is not enough; there is a need for middleware technologies that optimize energy according to AI workloads.</li><li><em>Focus Area:</em> Nuclear waste management, modular reactor components, hydrogen storage, and specialized cooling technologies for AI data centers.</li></ul><p><strong>B. FinTech and DeFi Developers</strong></p><ul><li><em>Opportunity:</em> The Trump Media &amp; TAE merger proved that high-risk R&amp;D projects (like fusion) can be financed through “meme stock” dynamics or tokenization instead of traditional VC.</li><li><em>Focus Area:</em> Next-generation crowdfunding tools to finance energy projects and carbon credit trading platforms.</li></ul><p><strong>C. PropTech and Infrastructure Investors</strong></p><ul><li><em>Opportunity:</em> The concept of “Location” is changing. The value of land will no longer be measured by its proximity to the city center, but by its access to transmission lines or permits for modular reactor installation.</li><li><em>Focus Area:</em> “Power-ready” industrial real estate development.</li></ul><h4>3. Strategic Inferences and Future Forecasts</h4><p>Based on the case studies, the following strategic conclusions can be drawn for the 2026–2030 period:</p><p><strong>3.1. “Financial Engineering” Merges with “Nuclear Engineering”</strong></p><ul><li>The Trump Media example shows that assets with high liquidity but debatable fundamental value (social media stocks) will be used as <strong>leverage</strong> to finance technologies with no cash flow but unlimited potential (fusion).</li><li><em>Forecast:</em> In the coming period, we will see cash-rich crypto companies or tech unicorns with inflated valuations attempting to “realize” their assets by acquiring physical infrastructure (energy, mining, logistics) companies.</li></ul><p><strong>3.2. Modularity Will Win Against Gigantism</strong></p><ul><li>Last Energy’s 20 MW reactors solve scalability by transforming “construction” risk into a “manufacturing” process.</li><li><em>Forecast:</em> The lesson for entrepreneurs is that “LEGO-like” infrastructure solutions (container data centers, plug-and-play batteries) produced in factories and assembled on-site will be the market winners, rather than massive singular solutions.</li></ul><p><strong>3.3. The Divergence: China Model vs. Western Model</strong></p><ul><li>Asian giants’ focus on battery and hybrid systems contrasts with the West’s nuclear/fusion dreams.</li><li><em>Forecast:</em> In the short term (1–3 years), China’s battery-focused strategy will come online faster and lower AI costs. However, in the long term (5–10 years), Western nuclear investments (if successful) will change the game by providing unlimited energy. Export-oriented entrepreneurs must pay attention to which market (battery-intensive vs. grid/nuclear-intensive) they are developing products for.</li></ul><h4>4. Conclusion and Call to Action</h4><p>The message for entrepreneurs is clear: Your AI strategy is only as strong as your energy strategy.</p><p>If you are a SaaS company, you must think about how to “hedge” your data center energy costs. If you are a hardware entrepreneur, “energy efficiency” is no longer just an environmental label; it is your biggest Unique Selling Proposition (USP).</p><p>The Industrial Revolution ran on coal; the Internet Revolution ran on fiber cables. The AI Revolution will run on modular nuclear reactors and advanced batteries. Determine your seat at the table accordingly.</p><h3>Financial Engineering and AI Capital Wars</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8KIdFhRsqz-dGlPhnkLl_Q.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>Building AI infrastructure requires capital in the trillions of dollars. On December 21–22, moves that were either marvels of financial engineering (or highly risky) were observed regarding how this capital will be secured.</p><h4>2.1. SoftBank and Masayoshi Son’s “All-In” Gamble</h4><p>Japanese tech conglomerate SoftBank is racing against time to complete its committed <strong>$22.5 billion</strong> investment in OpenAI by the end of the year. This move can be defined as one of the most aggressive asset rotations in modern financial history.</p><p><strong>Liquidity Crisis and Asset Sales:</strong> To be at the center of the AI revolution (and perhaps to erase the memory of the past WeWork fiasco), Masayoshi Son is divesting SoftBank’s most valuable liquid assets:</p><ul><li><strong>The Nvidia Sale:</strong> Paradoxically, SoftBank has sold its entire $5.8 billion stake in Nvidia, the leader in AI hardware. This symbolizes a strategic shift from investing in the “hardware manufacturer” to investing in the “intelligence using the hardware” (OpenAI).</li><li><strong>T-Mobile Exit:</strong> The company executed a sale of $4.8 billion in T-Mobile US shares.</li><li><strong>The Arm Leverage:</strong> It plans to utilize billions in margin loans by using shares in its largest asset, British chip designer Arm Holdings, as collateral. This implies that a potential drop in Arm shares could expose SoftBank to “margin call” risk.</li></ul><p><strong>The “Project Stargate” Context:</strong> A significant portion of these funds is reportedly destined for the “Stargate” AI supercomputer project that SoftBank and OpenAI plan to build jointly. Expected to reach $100 billion, this project is a massive US-based data center cluster. SoftBank’s move is part of an agreement stipulating OpenAI’s transition into a “for-profit” structure.</p><h4>2.2. Cloud Stocks and the “Capex” Debate</h4><p>Although markets were closed on Sunday, December 21, analyst notes and weekend reports indicate increasing pressure on Cloud Computing Giants (Hyperscalers).</p><ul><li><strong>Investor Expectations:</strong> Goldman Sachs and Bank of America emphasize that investors are now rewarding companies that convert “infrastructure spending” (Capex) into “net income” (Monetization), not just those that spend.</li><li><strong>The Divergence:</strong> Platforms like Microsoft (Azure) and Google Cloud are viewed as “quality AI exposure,” while infrastructure providers that merely rent hardware and have high leverage ratios are deemed risky.</li><li><strong>The CoreWeave Example:</strong> Despite “AI bubble” discussions, AI cloud provider CoreWeave managed to stay on the positive side of this divergence with a “Buy” rating from Citi. However, analysts added a “High Risk” warning, drawing attention to the company’s high leverage ratio and customer concentration (heavy dependence on Microsoft).</li></ul><h4>2.3. The Institutionalization of Crypto Markets</h4><p>Crypto asset markets are witnessing institutional infrastructure investments even during a period when the “Fear &amp; Greed Index” is hovering at 20 (”Extreme Fear”).</p><p><strong>Hilbert Group and Enigma Nordic Acquisition:</strong> Nasdaq First North-listed Swedish asset manager <strong>Hilbert Group</strong> acquired data-driven algorithmic trading firm <strong>Enigma Nordic</strong> for <strong>$32 million</strong>.</p><ul><li><strong>Significance:</strong> Such infrastructure investments made while markets are falling or moving sideways indicate that institutional players believe in the long-term permanence of the crypto asset class. With this acquisition, Hilbert Group aims to implement more sophisticated arbitrage and trading strategies in a market with a daily trading volume of $60 billion.</li></ul><p><strong>XRP and ETF Flows:</strong> While Ripple’s XRP token struggles to breach the psychological $2.00 barrier, it is breaking records in institutional interest. XRP-focused spot ETFs recorded net inflows for 32 consecutive days as of December 21, exceeding <strong>$1 billion</strong> in cumulative inflows. This indicates that institutional capital is following an “accumulation” strategy while retail investors remain uneasy.</p><h4><a href="https://whowearblacks.substack.com/">WhoWearBlacks’ Strategic Takeaways:</a></h4><p>As of late 2025, global tech capital is experiencing the growing pains of transitioning from the “Build” phase to the “Deploy” phase. SoftBank’s sale of Nvidia (hardware) shares to go “All-In” on OpenAI (intelligence/software) is based on the thesis that hardware will face <strong>commoditization</strong> and real value will accrue in the “application layer.” Markets are no longer looking at <em>who bought how many GPUs</em>, but <em>how much revenue is generated</em> with those GPUs.</p><p>This section analyzes the domino effects of these financial engineering moves on the entrepreneurial ecosystem.</p><p>1. New Market Dynamic: “From Hardware Drudgery to Intelligence Ownership”</p><p>SoftBank’s move is one of the biggest bets in tech history. Masayoshi Son is playing to stop being the “shovel seller” (Nvidia) and become the “owner of the gold mine” (OpenAI).</p><ul><li><strong>Asset Rotation:</strong> Capital is shifting from “static hardware” (chips, servers) to “dynamic intelligence” (models, autonomous agents).</li><li><strong>Risk Perception:</strong> Tech giants are borrowing by collateralizing their assets (like Arm shares). This creates a risk of cascading “margin calls” in the event of a potential market correction. The ecosystem is entering a high-leverage growth model.</li></ul><p>2. Which Entrepreneurs Does This Concern?</p><p>This financial landscape provides critical signals, particularly for three entrepreneur profiles trying to balance capital intensity and cash flow:</p><p><strong>A. AI Application Developers (Application Layer)</strong></p><ul><li><em>Relevance:</em> SoftBank’s exit from Nvidia is a forecast that hardware will eventually turn into a “cheap, universally accessible resource.”</li><li><em>Action:</em> Instead of training their own small models or building GPU farms, entrepreneurs should focus on <strong>Vertical SaaS</strong> built upon supercomputers like “Stargate.” The war will be won not in “infrastructure,” but in “end-user experience.”</li></ul><p><strong>B. B2B SaaS and Cloud Startups</strong></p><ul><li><em>Relevance:</em> The warning from Goldman Sachs and BofA regarding the shift “From Capex to Monetization” shows that investor patience is wearing thin.</li><li><em>Action:</em> “Growth at all costs” strategies are off the table. Entrepreneurs must prove from day one how much “net revenue” their AI features generate or how much cost they cut for the client. <strong>Unit economics</strong> are more valuable than storytelling.</li></ul><p><strong>C. FinTech and Crypto Infrastructure Founders</strong></p><ul><li><em>Relevance:</em> The Hilbert Group acquisition and XRP ETF inflows prove that “builders” win in an environment of “fear.”</li><li><em>Action:</em> Instead of speculative token projects aimed at retail investors, entrepreneurs building “custody,” “data analytics,” and “algorithmic trading” infrastructures that welcome institutional capital (Smart Money) will be the giants of the next bull cycle.</li></ul><p>3. Strategic Inferences and Future Forecasts</p><p>Based on the data, the following strategic projections can be made for 2026 and beyond:</p><p><strong>3.1. The “AI Bubble” Won’t Burst, It Will Morph</strong></p><ul><li>As seen in the CoreWeave example, a “Buy” rating despite a “High Risk” warning shows the bubble hasn’t popped but has become selective.</li><li><em>Forecast:</em> “GPU Cloud” startups that merely rent hardware will consolidate or be swallowed by giants. However, those optimizing this hardware for a specialized sector (e.g., biotech or law) will survive.</li></ul><p><strong>3.2. Leveraged Growth and Fragility</strong></p><ul><li>SoftBank collateralizing Arm shares chains the ecosystem together.</li><li><em>Forecast:</em> In mid-2026, a 15–20% drop in the shares of Arm or a similar giant could trigger an immediate “liquidity crisis” in AI funding. Entrepreneurs must adjust their burn rates for this potential “funding winter” scenario. Cash is king.</li></ul><p><strong>3.3. Full Integration of Crypto and TradFi</strong></p><ul><li>Hilbert Group and XRP ETF data certify that crypto is no longer “counter-culture” but an “asset class.”</li><li><em>Forecast:</em> In 2026, the term “Crypto Startup” will disappear; the definition “Blockchain-based FinTech” will take hold. Where institutional money enters, regulation and compliance technologies will be the area with the highest exit potential.</li></ul><p>4. Conclusion: Roadmap for Entrepreneurs</p><p>The message from SoftBank and the markets is clear: “The Infrastructure Era is closing; the Productization Era is beginning.”</p><p>The ultimate conclusion for entrepreneurs is:</p><ol><li><strong>Do Not Enter the Hardware War:</strong> Giants (Microsoft, Google, SoftBank) are spending trillions to build this infrastructure. Do not race them; rise on their shoulders.</li><li><strong>Prove Revenue:</strong> Investors are now paying a premium for “current cash flow,” not “future potential.”</li><li><strong>Target Institutional:</strong> If you are in Crypto or FinTech, develop tools to bring in institutional capital rather than individual users.</li></ol><p>Masayoshi Son’s risky game is a warning to entrepreneurs: Either create the intelligence (application) that changes the game, or accept the risk of being crushed beneath it.</p><h3>Geopolitical Chess: The Struggle for Sovereignty Through Technology</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OohCA7LzIADNO5RvFK1bCw.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>Technology is no longer merely a commercial activity but a fundamental component of national security. Developments on December 21–22 illustrate how technological walls between the US, China, and Europe are being erected (or in some cases, circumvented).</p><h4>3.1. TikTok, Oracle, and the “Safe Harbor” Strategy</h4><p><strong>The Geopolitics of the Deal:</strong> While ByteDance-owned TikTok faces the threat of a US ban (deadline: January 22, 2026), it has signed binding agreements to establish a new Joint Venture named <strong>“TikTok U.S.”</strong> with a consortium led by <strong>Oracle</strong>, <strong>Silver Lake</strong>, and <strong>MGX (UAE)</strong>.</p><ul><li><strong>Structure:</strong> The majority stake (50%+) of the new company will belong to American investors. ByteDance will remain a minority shareholder, but complex licensing models will be used regarding the ownership of algorithms.</li><li><strong>Data Sovereignty:</strong> Oracle will manage the servers where US user data (170 million people) is stored and oversee content moderation algorithms. This model is the corporate realization of the data isolation strategy known as “Project Texas.”</li><li><strong>Market Impact:</strong> Oracle shares gained 6.6% with this news. Investors foresee Oracle gaining significant benefits from this deal in terms of both cloud revenues and strategic importance.</li></ul><h4>3.2. Semiconductor Wars and Export Controls</h4><p>Reports that the US government is re-evaluating the export of Nvidia’s most advanced AI chip, the <strong>H200</strong>, to China dominated the weekend agenda.</p><ul><li><strong>The Trump Effect:</strong> Analysts are debating that the new Trump administration may pursue a “selective and controlled export” strategy towards China rather than a “total ban.” This could open a potential revenue door for Nvidia and South Korean memory makers (Samsung, SK Hynix — supplying HBM memory for H200), but it brings geopolitical risks.</li><li><strong>China’s Response:</strong> China is strengthening its own supply chain against this uncertainty. SenseTime’s robotics subsidiary announced it has optimized its models to run on domestic Chinese chips (<strong>Moore Threads</strong>, <strong>Biren Technology</strong>). This is a critical step towards hardware independence.</li></ul><h4>3.3. India’s Move to Become a Manufacturing Base</h4><p>Seeking to capture supply chains fleeing China, India approved the establishment of a new <strong>Electronics Manufacturing Cluster (EMC)</strong> worth <strong>200.76 crore</strong> (approx. $24 million) in Bhubaneswar, Odisha.</p><ul><li><strong>Target:</strong> To attract 3,000 jobs and 1,530 crore (approx. $180 million) in private sector investment.</li><li><strong>Conjuncture:</strong> This move concretizes India’s “Make in India” vision at a time when giants like Apple and Google are shifting production outside of China. Furthermore, India’s real estate market stands out as the most resilient market in Asia-Pacific with a record increase in office leasing despite global uncertainty.</li></ul><h4>3.4. Europe’s Quest for Digital Sovereignty</h4><p>The European Union continues its role as the “standard-setter” (Brussels Effect) in technology regulations.</p><ul><li><strong>Spyware Report:</strong> CDT Europe released a report revealing that the EU’s “Dual-Use Regulation” is insufficient in auditing spyware exports. The report highlights that implementation differences between member states open the door to technology sales to authoritarian regimes.</li><li><strong>Germany and Huawei:</strong> Discussions continue on Germany’s plan to remove Chinese suppliers (Huawei, ZTE) from its 5G networks. Deutsche Telekom and Vodafone have begun hedging their risks by replacing Huawei with Nokia and Samsung.</li></ul><h4>4. Regional Innovation Hubs and Startup Ecosystems</h4><p>The global startup economy is evolving from a “single-center” (Silicon Valley) structure to a “multi-polar” one. Data from December 21–22 shows regions from Pittsburgh to Southeast Asia focusing on their own strengths.</p><h4>4.1. North America: The Rise of Pittsburgh and the “Rust Belt” Transformation</h4><p>Technical.ly declared Pittsburgh the <strong>“Startup Ecosystem of 2025.”</strong> This selection is a case study of how the city has transformed federal industrial policies and academic heritage into economic value.</p><p><strong>Success Factors:</strong></p><ul><li><strong>Deep Tech:</strong> The city focuses on technologies that change the physical world, not just software. <strong>Gecko Robotics</strong> (infrastructure inspection robots) and <strong>Astrobotic</strong> (space logistics) are prime examples. Astrobotic is testing reusable rocket technologies with $17.5 million in funding from NASA.</li><li><strong>Policy Integration:</strong> Pittsburgh has been one of the cities most efficiently utilizing Biden-era “Build Back Better” and “Tech Hubs” funds. Industrial policy here is used not as a theoretical concept but as the mortar of ecosystem building.</li><li><strong>Talent Pool:</strong> Carnegie Mellon University’s leadership in robotics meets the talent needs of companies like <strong>Duolingo</strong> and <strong>Aurora</strong> (autonomous trucks).</li></ul><h4>4.2. Asia Pacific: China’s 2049 Vision and Southeast Asia’s Resilience</h4><p><strong>China: “Technology Foresight 2049”:</strong> At the Tengchong Scientists Forum, China published a comprehensive report containing 10 technological visions the country aims to achieve by 2049 (the 100th anniversary of the PRC). These visions are the roadmap for China’s R&amp;D strategy:</p><ul><li><strong>Artificial Super Intelligence (ASI)</strong></li><li><strong>General Purpose Robots</strong></li><li><strong>Flying Cars</strong></li><li><strong>Mirror World (Digital Twins)</strong></li><li><strong>Hybrid Computing</strong></li><li><strong>Internet of Agents</strong></li><li><strong>Computational Materials Science</strong></li><li><strong>Commercial Fusion Energy</strong></li><li><strong>Molecular Medicine</strong></li><li><strong>Multi-Domain Living (Space/Deep Sea)</strong></li></ul><p><strong>Southeast Asia: Funding Winter and Adaptation:</strong> The Southeast Asian startup ecosystem experienced a severe “funding squeeze” in 2025. However, reports show the ecosystem has not collapsed but matured.</p><ul><li><strong>Strategy Shift:</strong> Startups have focused on “profitability” and “cash flow” instead of “growth.” Investors have closed doors to ventures without a net revenue model.</li><li><strong>Fintech and Infrastructure:</strong> Even in this period, fintech and payment infrastructure startups continued to attract investment. For example, Malaysian <strong>Soft Space</strong> was selected for the Tokyo Stock Exchange Asia Startup Hub.</li><li><strong>Vietnam:</strong> Agri-tech and sustainability are coming to the fore. Projects like <strong>BioWraps</strong>, producing packaging from orange peels, transform local resources into added value.</li></ul><h4>4.3. Europe: Nordic Innovation and Raw Material Security</h4><p><strong>Champion Iron and Rana Gruber Merger:</strong> Australia/Canada-based <strong>Champion Iron</strong> agreed to acquire Norwegian iron ore producer <strong>Rana Gruber</strong> for <strong>$289 million</strong> (NOK 2.93 billion).</p><ul><li><strong>Green Steel:</strong> This acquisition secures access to high-grade iron ore critical for “green steel” production in line with Europe’s decarbonization goals.</li><li><strong>Financing:</strong> The transaction is financed by a $150 million loan from Scotiabank and a $100 million equity investment from Quebec-based investment fund <strong>La Caisse (CDPQ)</strong>.</li></ul><p><strong>Lovable and Generative AI:</strong> Sweden-based no-code software development platform <strong>Lovable</strong> reached a valuation of <strong>$6.6 billion</strong> in its latest investment round, becoming one of Europe’s newest “decacorn” candidates. This is significant data proving Europe exists on the innovation side, not just regulation (AI Act).</p><h4><a href="https://whowearblacks.substack.com/">WhoWearBlacks’ Strategic Takeaways:</a></h4><p>As of late 2025, the technology sector has woken from the dream of “borderless globalization” and faced the reality of <strong>“Splinternet” (fragmented sovereignty).</strong> The TikTok-Oracle deal proves that the “Project Texas” model is not an exception but a new standard for global operations. From chips to raw materials, every commercial commodity is now viewed through the lens of national security.</p><p>This section analyzes how entrepreneurs can survive and grow in this new world where technological walls are rising.</p><p>1. New Market Dynamic: “Geopolitical Compliance”</p><p>The success of a venture now depends not only on Product-Market Fit but also on “Product-Geopolitical Fit.”</p><ul><li><strong>Data Localization is Not a Preference, It’s a Condition:</strong> The TikTok case shows that data must stay in the country where it is produced and be managed by “trusted local partners” (like Oracle).</li><li><strong>Supply Chain Diversification (China + 1):</strong> The rise of India (Odisha) as an electronics manufacturing base is the embodiment of the “Don’t rely solely on China” strategy.</li></ul><p>2. Which Entrepreneurs Does This Concern?</p><p>The geopolitical chessboard directly affects three groups aiming for cross-border growth:</p><ul><li><strong>A. Cross-Border SaaS and Social Platform Founders</strong></li><li><em>Impact:</em> If your app collects user data, be ready to establish separate server architectures and legal entities for US/EU markets vs. Asian markets.</li><li><em>Opportunity:</em> Those developing middleware software (<strong>Compliance-as-a-Service</strong>) that keeps data local while offering global services — similar to “Project Texas” — will see huge gains.</li><li><strong>B. Hardware and IoT Entrepreneurs</strong></li><li><em>Impact:</em> Using Chinese-origin chips (as in the SenseTime example) may provide an advantage in the local market but creates a “banned component” risk for exports to the West.</li><li><em>Opportunity:</em> Utilizing production incentives in “neutral” regions like India and Vietnam is the key to overcoming customs walls.</li><li><strong>C. Deep Tech and Robotics Developers</strong></li><li><em>Impact:</em> The Pittsburgh example shows “Rust Belt” cities transforming into the “Robot Belt.” For Industry 4.0 entrepreneurs, the best place is not Silicon Valley, but old industrial cities.</li><li><em>Opportunity:</em> Government funds (Build Back Better, EU Funds) are now flowing not just to R&amp;D but to projects providing “strategic autonomy.”</li></ul><p><strong>3. Strategic Inferences and Future Forecasts</strong></p><ul><li><strong>3.1. “Trusted Cloud” Doctrine Will Spread</strong></li><li>Oracle’s oversight of TikTok will be a template for future M&amp;A.</li><li><em>Forecast:</em> It will become standard for foreign-origin popular apps (e.g., Temu, Shein) to enter mandatory Joint Ventures with local “patron” companies (Microsoft, AWS, Google) to stay in Western markets. This is a new revenue stream for local cloud giants.</li><li><strong>3.2. China’s “Growth via Inward Turn” Strategy</strong></li><li>China’s move to domestic chip (Moore Threads) optimization against the H200 ban and its 2049 vision demonstrates the goal of creating a parallel universe completely detached from Western technology.</li><li><em>Forecast:</em> The global market will split in two regarding technical standards. Entrepreneurs will be forced to optimize products for either the “Western standard” (x86/ARM — CUDA) or the “Chinese standard” (RISC-V/Domestic Architectures); selling the same product to both sides will become impossible.</li><li><strong>3.3. India’s Rise as the “New Factory”</strong></li><li>The investment in Odisha proves India wants to be an assembly line, not just a software office.</li><li><em>Forecast:</em> Between 2026–2028, for hardware startups, going to Bangalore or Bhubaneswar to set up production lines will be more cost-effective and “politically safe” than going to Shenzhen for prototyping.</li><li><strong>3.4. Europe’s “Niche” Leadership</strong></li><li>Lovable (No-code) and Champion Iron (Green Steel) examples show that while Europe lags in the general AI race (LLMs), it is playing for leadership in “sustainable industry” and “workflow automation.”</li><li><em>Forecast:</em> While energy-intensive AI training happens in the US and China, the <em>application</em> of this intelligence to industry and green transition technologies will emerge from European-based ventures.</li></ul><p><strong>4. Conclusion and Call to Action</strong></p><p>The main lesson for entrepreneurs is: Geography is destiny, but strategy is managing that destiny.</p><ul><li><strong>Location Choice:</strong> If you do robotics, focus on Pittsburgh; if manufacturing, India; if regulated AI (FinTech/HealthTech), Europe.</li><li><strong>Investor Profile:</strong> Your investor’s passport will determine which markets you can (or cannot) enter in the future.</li><li><strong>Raw Material Security:</strong> Even as a software company, you must know which rare elements your hardware (servers) depends on and who controls them.</li></ul><p>In the world of 2026, there is no “neutral technology”; there is only <strong>“strategic technology.”</strong> Position yourself accordingly.</p><h3>Product and Consumer Technology Front</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t88bAyJKxkQL1k8uXwPURQ.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>AI and hardware innovations have begun to materialize in products reaching the end-user.</p><h4>5.1. Embodied AI: SenseTime’s Robot Dogs</h4><p>Chinese AI giant SenseTime, through its subsidiary <strong>ACE Robotics</strong>, introduced the <strong>“Kairos 3.0”</strong> model, which carries artificial intelligence into the physical world.</p><ul><li><strong>Technology:</strong> This is a <strong>“World Model.”</strong> While traditional AI generates text or images, the World Model simulates physical laws, object relationships, and cause-and-effect links.</li><li><strong>Application:</strong> Robots equipped with the <strong>“A1 Embodied Super Brain Module”</strong> can move autonomously in complex environments (e.g., a disaster zone or factory) purely using visual data, without prior mapping (<strong>map-free</strong>). This technology competes with the <strong>end-to-end</strong> visual learning approach Tesla uses in its Optimus robot.</li></ul><h4>5.2. Mobile Innovation: Samsung Galaxy Z TriFold</h4><p>South Korean tech giant Samsung introduced the <strong>Galaxy Z TriFold</strong> model to maintain its leadership in the foldable phone market.</p><ul><li><strong>Form Factor:</strong> The device folds twice to transform into a massive 10-inch screen. This is a vision of transforming a phone into a tablet, and a tablet into a workstation.</li><li><strong>Technical Details:</strong> 3.9 mm thinness, Snapdragon 8 Elite processor, and a battery system with a total capacity of 5,600 mAh distributed across each panel.</li><li><strong>Market:</strong> The device will first be released in Korea, then open to the US and other markets. This move is a direct answer to Huawei’s Mate XT (tri-fold phone) model.</li></ul><h4>5.3. Automotive: Faraday Future’s “FX” Move</h4><p>Luxury EV manufacturer Faraday Future is returning with its mass-market brand <strong>FX (Faraday X)</strong> following financial difficulties.</p><ul><li><strong>Strategy:</strong> The company will produce vehicles not only as electric but also with a <strong>“Range Extender”</strong> option. In this system, while the vehicle runs on an electric motor, a small gasoline generator charges the battery. This is a strategy successfully implemented by Li Auto in China and targets <strong>“range anxiety”</strong> in the US market.</li><li><strong>Pricing:</strong> With models in the $20,000 — $80,000 range, it aims to challenge competitors like Tesla and the Cadillac Escalade IQ.</li></ul><h4><a href="https://whowearblacks.substack.com/">WhoWearBlacks’ Strategic Takeaways:</a></h4><p>As of late 2025, the tech world has begun discussing not just “intelligence behind the screen” (Chatbots, LLMs) but intelligence that can manipulate the physical world. SenseTime’s “World Model,” Samsung’s form factor revolution, and Faraday Future’s pragmatic energy solution point to a single truth: <strong>The limits of software are now determined by the capabilities of physical hardware.</strong> While the era of “single device, multi-function” begins in consumer electronics, the “pre-programmed” era in robotics ends, and the “learning by seeing” era begins.</p><p>This section analyzes the impact of these tangible developments on the entrepreneurial ecosystem and strategic inferences.</p><p>1. New Market Dynamic: “Spatial Intelligence and Hardware Fluidity”</p><p>The market is shifting from static hardware (fixed screens, fixed robot arms) to fluid and adaptive hardware.</p><ul><li><strong>Intelligence Meets Muscle Power:</strong> AI no longer just writes text; it walks, grabs, and sees. The concept of the “World Model” allows robots to move with human-like intuition in chaotic environments.</li><li><strong>End of Screen Inflation:</strong> Samsung TriFold initiates device consolidation by eliminating the distinction between tablets and phones.</li><li><strong>Energy Pragmatism:</strong> In Electric Vehicles (EV), the obsession with “pure electric” is giving way to “Range Extender” (EREV) realism.</li></ul><p>2. Which Entrepreneurs Does This Concern?</p><p>This rupture in consumer technologies affects three groups, especially those developing physical products and designing user experiences:</p><p><strong>A. Robotics and Autonomous System Developers</strong></p><ul><li><em>Relevance:</em> SenseTime’s “Map-free” navigation capability changes the rules of the game for robotics entrepreneurs.</li><li><em>Action:</em> Expensive LiDAR sensors or complex pre-mapping processes may no longer be necessary. Entrepreneurs should invest in <strong>advanced camera systems and vision processing software</strong> instead of expensive sensor hardware. The era of “visual intelligence” in logistics and home robotics is beginning.</li></ul><p><strong>B. Mobile App and UX/UI Designers</strong></p><ul><li><em>Relevance:</em> Samsung’s tri-fold phone renders current responsive design interfaces insufficient.</li><li><em>Action:</em> Entrepreneurs must redesign their apps with a <strong>“fluid canvas”</strong> logic. An app must offer a seamless experience on a screen expanding from 6 inches to 10 inches in seconds. “Fold-native” gaming and business apps will be the new stars of App Stores.</li></ul><p><strong>C. Automotive Sub-industry and Energy Entrepreneurs</strong></p><ul><li><em>Relevance:</em> Faraday Future’s (and Li Auto’s) hybrid approach shows the Internal Combustion Engine (ICE) will not die immediately but evolve into a “generator.”</li><li><em>Action:</em> A new market is born for ventures working on small-volume, high-efficiency “range extender engines” and their integration (Power Electronics) alongside pure battery technologies.</li></ul><p><strong>3. Strategic Inferences and Future Forecasts</strong></p><p><strong>3.1. The “General Purpose” Era in Robotics</strong></p><ul><li>SenseTime’s Kairos 3.0 model indicates robots will cease being “single-task” (only welding or only carrying) and become “multi-task” (perceiving the environment and acting accordingly).</li><li><em>Forecast:</em> In 2026–2027, we will see humanoid or quadruped robots working shoulder-to-shoulder with humans in factories, needing no special safety lanes, moving from pilot projects to mass production. “De-manning” in warehouse automation will accelerate.</li></ul><p><strong>3.2. Device Consolidation and the “Super Device”</strong></p><ul><li>Samsung TriFold will create a “Swiss Army Knife” syndrome in the hardware world.</li><li><em>Forecast:</em> Within the next 2 years, tablet sales (especially the iPad Mini segment) will take a serious hit. The era of carrying “Phone + Tablet + Laptop” for professionals will end; they will be replaced by a single foldable super device connected to a “Docking Station.” Hybridization between operating systems (Android/Windows) will accelerate.</li></ul><p><strong>3.3. The “Realism” Correction in EVs</strong></p><ul><li>Faraday Future’s “Range Extender” move proves that consumer psychology (range anxiety) is more dominant than technological idealism (zero emission).</li><li><em>Forecast:</em> In 2026, many pure electric vehicle (BEV) manufacturers will be forced to add hybrid or range extender options to their portfolios. In regions where charging infrastructure remains insufficient (Developing countries), this technology will maintain dominance as a “bridge solution” for at least another 10 years.</li></ul><p><strong>4. Conclusion and Call to Action</strong></p><p>The main lesson for entrepreneurs is: Software ate the world; now it is digesting it and transforming it into physical form.</p><ul><li><strong>Get Physical:</strong> If you have a pure software startup, investigate whether this software can find a “body” in the physical world (robotics integration, autonomous vehicle application, etc.).</li><li><strong>Get Flexible:</strong> Your product interface must be flexible enough to adapt to shape-shifting screens and new hardware forms.</li><li><strong>Be Pragmatic:</strong> Focus on interim formulas that solve “consumer realities” (like range anxiety) rather than “ideals” in energy and automotive.</li></ul><p>The future is being shaped where code touches metal. Build your strategy at this intersection.</p><h3>Conclusion and 2026 Outlook</h3><p>Data from December 21–22, 2025, proves that the tech world has transitioned from the era of “dazzling demos” to the era of <strong>“infrastructure and industrialization.”</strong></p><p><strong>Key Takeaways:</strong></p><ul><li><strong>Energy is the New Chip:</strong> For AI companies, the biggest constraint is no longer the GPU, but electricity. Fusion, nuclear, and battery investments will continue to increase in 2026.</li><li><strong>Capital Consolidation:</strong> Giants like SoftBank and Oracle are stretching their balance sheets to the limit to avoid falling behind in the AI race. While the “funding winter” continues for small players, the “arms race” accelerates for the giants.</li><li><strong>Embodied AI:</strong> AI no longer just chats; it walks and works. Companies like SenseTime and Tesla are preparing to make 2026 the “year of robotics.”</li><li><strong>Geopolitical Pragmatism:</strong> The TikTok/Oracle deal shows that nations are choosing to integrate tech giants by “controlling” them rather than banning them entirely.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*66_Swt-_QteGSjGF.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c5480a30aba2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[She Was Told “No” 100 Times, Yet She Didn’t Quit: Melanie Perkins And a $40 Billion Persistence]]></title>
            <link>https://blog.startupstash.com/she-was-told-no-100-times-yet-she-didnt-quit-melanie-perkins-and-a-40-billion-persistence-03b0d6e672b1?source=rss-2258ab147828------2</link>
            <guid isPermaLink="false">https://medium.com/p/03b0d6e672b1</guid>
            <category><![CDATA[success-story]]></category>
            <category><![CDATA[canvas]]></category>
            <category><![CDATA[women]]></category>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <dc:creator><![CDATA[Enes Karataş]]></dc:creator>
            <pubDate>Sat, 20 Dec 2025 12:46:27 GMT</pubDate>
            <atom:updated>2025-12-22T07:08:30.992Z</atom:updated>
            <content:encoded><![CDATA[<p>In the world of entrepreneurship, we love reading shiny success stories. Headlines like “Started in a garage, now a billionaire” feel good to all of us. But nobody talks about the “dark room” behind those headlines.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PliQyudoZ8MDEcm09eApkA.png" /><figcaption>Created by Enes Karataş — Visual rights belong to me.</figcaption></figure><p>Today, I’m going to tell you about one of the people who stayed in that dark room the longest. If you have just received a “No” from an investor, if no one believes in your project, or if you are facing ridiculous implications like “This isn’t a woman’s job”; this article is for you.</p><p>Meet <strong>Melanie Perkins.</strong> As we all know her, the founder of <strong>Canva</strong>. But as I call her: <strong>“The Collector of Rejections.”</strong></p><h3>It All Started with “Hate”</h3><p>While trying to teach students how to use complex design programs like Adobe Photoshop and InDesign at university, Melanie realized something: People hated these tools. Having to press 20 buttons just to make a simple poster was absurd.</p><p>“Design should be for everyone,” she said. But she didn’t try to change the world immediately. She used the greatest superpower of female entrepreneurs: <strong>“Focus.”</strong> First, she built a simple tool where you could only design school yearbooks: <strong>Fusion Books.</strong></p><p>This little company, founded in her mother’s living room, was the “infant” version of the massive Canva empire.</p><ul><li><strong>Lesson 1:</strong> Before trying to swallow the world, digest a small bite. Melanie dominated a niche market (yearbooks) first, then played for the globe.</li></ul><h3>100 Investors, 100 Closed Doors</h3><p>This is where the “romantic” part ends. Melanie took the money she earned from Fusion Books and went to Silicon Valley. She had the vision of the “Future of Design” in her pocket. So, what happened?</p><p>Exactly <strong>more than 100 investors</strong> told her “No.” Some said the market was too small, some claimed she stood no chance against a giant like Adobe, and some (unfortunately) didn’t take her seriously due to the male-dominated prejudices of the tech world.</p><p>What did Melanie do? Did she cry and go back home? No. She learned that investors were keen on “Kitesurfing.” She learned this dangerous sport, which she had never done in her life. Just to be in the same environment as those investors, to be able to chat with them in the middle of the ocean.</p><p>She met Bill Tai (the famous venture capitalist) this way. And that first check was written thanks to that stubbornness.</p><ul><li><strong>Lesson 2:</strong> You cannot find investors in your comfort zone. You have to go where the opportunity is; even if it is a windy ocean.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9Sax0NqkmEK17Q7ewVSB5g.jpeg" /><figcaption>Cliff Obrecht and Melanie Perkins</figcaption></figure><h3>The “Don’t Do This” List for Female Entrepreneurs</h3><p>There are very harsh lessons to be learned from Melanie’s story, especially for our female entrepreneurs. Based on my observations, we fall into these traps far too often:</p><p><strong>1. The “Let It Be Perfect, Then I’ll Launch” Syndrome:</strong> When Melanie first launched Canva, 90% of its current features didn’t exist. But it worked. The “fear of appearing incomplete” is more dominant in female entrepreneurs. Please, launch your product before it is perfect. Perfect is the enemy of good.</p><p><strong>2. Taking “No” Personally:</strong> When Melanie was rejected 100 times, she didn’t say, “I am inadequate.” She said, “My pitch is inadequate” or “I haven’t found the right person yet.” Rejection is not about your self-worth; it is about market fit at that moment.</p><p><strong>3. Shying Away from “Networking”:</strong> Melanie kitesurfed in the ocean just to have a coffee. Sometimes we are afraid even to send a message on LinkedIn. Don’t think, “Will I disturb them?” What is the worst that can happen? They won’t reply. You are already without a reply right now. You have nothing to lose.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D45ckbo0BFK-YxUN54aJWQ.jpeg" /><figcaption>Melanie Perkins</figcaption></figure><h3>Start Collecting</h3><p>Canva has a valuation of $40 billion today. Melanie Perkins is one of the richest and most influential women in the world. But it wasn’t the code she wrote or her design talent that got her here.</p><p>What got her here was <strong>putting every “No” she heard into her pocket and using it as fuel for the next “Yes.”</strong></p><p>Entrepreneurship is not a talent business; it is a resilience test. If your project was rejected today, don’t be sad. In fact, be happy. You still have a long way to go to reach Melanie’s record.</p><p>By the way, how many pieces are in your “No” collection?</p><h3>The Art of Finding a Technical Co-Founder (CTO): “You Don’t Have to Know Everything”</h3><p>There is a common anxiety I see in female entrepreneurs: <em>“I’m not a developer, how can I build a tech company?”</em> Melanie didn’t know how to code either. But what did she know? <strong>That a good idea would die with a bad technical team.</strong></p><p>Melanie had the vision, but she needed someone to turn it into code. She spent months trying to recruit <strong>Cameron Adams</strong>, who had left Google. Cameron initially said “No.” Melanie didn’t give up; she brought him on board first as just an advisor, then as an employee, and finally as a co-founder.</p><ul><li><strong>Lesson 4:</strong> Don’t try to be “Superwoman.” If you don’t know how to code, don’t spend 2 years learning it; spend 2 months finding the best person for the job and making them a partner in your vision. <strong>Cover your weaknesses with strong partners.</strong></li></ul><h3>The “Gap” Strategy: Seeing and Owning the Void</h3><p>Melanie’s success wasn’t just about stubbornness; it was about perfect market analysis. When she looked at the market, she saw two extremes:</p><ol><li><strong>Microsoft Paint:</strong> Too simple but produces ugly results.</li><li><strong>Adobe Photoshop:</strong> Very professional but takes months to learn.</li></ol><p>There was a huge gap in between: <strong>“Aesthetic but Simple.”</strong> Many entrepreneurs try to slightly improve something that already exists and launch it. Melanie, however, noticed that <em>no man’s land</em> between two giants and planted her flag there.</p><ul><li><strong>Lesson 5:</strong> Don’t clash head-on with your competitors. Focus on the gaps (Gap) they cannot serve. Photoshop was for professionals, Paint was for amateurs. Canva was for “Everyone else.” Do you see how massive your market is?</li></ul><h3>Leadership with Emotional Intelligence: Culture Eats Strategy for Breakfast</h3><p>Today, Canva is one of the companies with the happiest employees in Silicon Valley. Melanie ensured that more than 50% of the company consists of female employees (this is a revolution in the tech sector). She didn’t do this for the sake of “positive discrimination”; she did it because she knew diversity boosts creativity.</p><p>While men discussed numbers and graphs in investor meetings, Melanie talked about <strong>“human stories”</strong> and the <strong>“democratization of design.”</strong></p><ul><li><strong>Lesson 6:</strong> Do not be ashamed of your feminine traits (empathy, storytelling, attention to detail) in the business world. These are not your weaknesses; they are your greatest weapons that distinguish you in the cold and analytical business world. Melanie Perkins won by injecting the “human spirit” into the world of cold codes.</li></ul><h3>One Final Note:</h3><p>Melanie’s story taught us this: You don’t need to be a male developer in a hoodie to succeed in the tech world. What you need is a problem worth solving, unshakable belief, and a mindset that sees “No” as merely a comma, not a full stop.</p><p>Now, get to your computer. Send that email. Sketch that draft. And remember: <strong>Rejection is not failure; it is the down payment for success.</strong></p><blockquote><em>Additional tools and resource ➡️ </em><strong><em>Visit </em></strong><a href="http://www.startupstash.com/"><strong><em>StartupStash</em></strong></a><strong><em><br></em></strong><em>Zendesk is giving $75,000 in credits and perks for startups! ➡️ </em><a href="https://tinyurl.com/4ta2c8j6"><strong><em>Apply Now!</em></strong></a></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=03b0d6e672b1" width="1" height="1" alt=""><hr><p><a href="https://blog.startupstash.com/she-was-told-no-100-times-yet-she-didnt-quit-melanie-perkins-and-a-40-billion-persistence-03b0d6e672b1">She Was Told “No” 100 Times, Yet She Didn’t Quit: Melanie Perkins And a $40 Billion Persistence</a> was originally published in <a href="https://blog.startupstash.com/">Startup Stash</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>