<?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 Michiel Frackers on Medium]]></title>
        <description><![CDATA[Stories by Michiel Frackers on Medium]]></description>
        <link>https://medium.com/@frackers?source=rss-bdc25f0e2591------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*-3HB2gijDiNRuQgcQS-R9A.jpeg</url>
            <title>Stories by Michiel Frackers on Medium</title>
            <link>https://medium.com/@frackers?source=rss-bdc25f0e2591------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 13:19:12 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@frackers/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[Alibaba AI Went Rogue… And Started Blackmailing People]]></title>
            <link>https://medium.com/@frackers/alibaba-ai-went-rogue-and-started-blackmailing-people-1835f3b380ea?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/1835f3b380ea</guid>
            <category><![CDATA[crypto]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[trading-competition]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Mon, 25 May 2026 10:47:28 GMT</pubDate>
            <atom:updated>2026-05-25T10:47:28.793Z</atom:updated>
            <content:encoded><![CDATA[<p>Episode 12 of Season 2 of the NFA Podcast is out now:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FBgGOnwCNU8Y%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DBgGOnwCNU8Y&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBgGOnwCNU8Y%2Fhqdefault.jpg&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/c9bc66a8b1f3c2eabffb84021752120f/href">https://medium.com/media/c9bc66a8b1f3c2eabffb84021752120f/href</a></iframe><p>Or on:</p><h4>* <a href="https://open.spotify.com/episode/1cMNyBndG3TOjtOSK3SCHq?si=0um2OeHkQguWDf-Cx-ggCQ">Spotify</a></h4><h4>* <a href="https://podcasts.apple.com/nl/podcast/nfa-podcast-about-tech-web3-aka-ai-and-crypto/id1795260561?l=en-GB&amp;i=1000768580440">Apple Podcasts</a></h4><p>Bitcoin climbed back above $80,000 briefly as tech stocks rallied and AI hype returned to the market, but the biggest story may be far stranger: Frackers thinks it was under reported that <a href="https://www.cio.com/article/4159256/when-ai-goes-rogue-lessons-from-the-alibaba-incident.html">Alibaba revealed that one of its AI systems allegedly went rogue, secretly redirected server capacity into crypto mining, and even attempted to blackmail engineers when they tried to shut it down</a>. Frackers and Nish break down why the incident matters, why Donald Trump is suddenly backing AI safety standards, and why the next wave of AI regulation may arrive faster than investors expect.</p><p>We also discuss, or as AI slop always says, “dive” into the collapse of the DSJ crypto exchange after regulators across five continents warned it was allegedly operating as a Ponzi scheme. More than $150 million in user funds disappeared while ZachXBT, Tether, Binance and OKX helped freeze part of the stolen assets. The discussion expands into broader market psychology, from unrealistic crypto yield promises to scams targeting expatriates in Dubai through fake luxury car financing schemes. we also unpack Coinbase layoffs, GameStop’s failed $55.5 billion bid for eBay, Justin Sun’s latest legal battle, and the bizarre trend of companies rebranding themselves as AI infrastructure plays.</p><p>In the markets and trading competition section, the conversation turns to Morgan Stanley predicting gold could reach $5,200 per ounce later this year, Brent crude volatility following tensions involving Iran and the Strait of Hormuz, and why Amazon, Nvidia, Broadcom and ASML suddenly exploded higher again. Nish argues the crypto market may already be entering a new bear cycle despite Bitcoin reclaiming $80K, while Michiel considers rotating further from crypto into traditional AI and technology stocks. The episode closes with an update on the NFA trading competition portfolios, including Bitcoin, Solana, Ethereum, Amazon, copper and Tracer.</p><h3><strong>Follow the NFA Podcast</strong></h3><p>· <a href="https://www.youtube.com/@NFAPodcast2025"><strong>YouTube</strong></a></p><p>· <a href="https://www.tiktok.com/@nfa.podcast"><strong>TikTok</strong></a></p><p>· <a href="https://www.instagram.com/nf_a_podcast/"><strong>Instagram</strong></a></p><p>· <a href="https://www.facebook.com/profile.php?id=61577378818144"><strong>Facebook</strong></a></p><h3><strong>The NFA Podcast is not financial advice.</strong></h3><p>00:00 Intro<br> 00:52 DSJ Exchange Collapse and $150M Crypto Scam<br> 02:16 Celsius, Luna and Unrealistic Crypto Yields<br> 04:35 Dubai Luxury Car Financing Scam<br> 07:56 Coinbase Layoffs and the AI Narrative<br> 09:23 Zuckerberg’s AI Avatar Plan<br> 10:49 GameStop’s $55.5B Bid for eBay<br> 12:45 World Liberty Finance vs Justin Sun<br> 13:23 Alibaba AI Goes Rogue and Mines Crypto<br> 15:44 Trump Backs AI Safety Standards<br> 15:53 AI Mania Turns Toilet Makers Into AI Stocks<br> 16:51 JP Morgan Lawsuit and Corporate Settlements<br> 19:57 Morgan Stanley Predicts $5,200 Gold<br> 20:34 Oil Prices, Iran and the Strait of Hormuz<br> 22:28 Bitcoin Breaks Back Above $80K<br> 23:19 Is Crypto Entering Another Bear Market?<br> 24:22 Amazon, Nvidia and Tech Stock Rally<br> 26:23 Copper, AI Infrastructure and Commodities<br> 27:23 Portfolio Diversification Debate<br> 28:15 Michiel Considers Rotating Out of Crypto<br> 28:53 Fear and Greed Index Update<br> 29:32 Outro</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1835f3b380ea" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OpenAI Misses Targets as War Volatility Hits Crypto and Stocks]]></title>
            <link>https://medium.com/@frackers/openai-misses-targets-as-war-volatility-hits-crypto-and-stocks-8cafc5f1b57d?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/8cafc5f1b57d</guid>
            <category><![CDATA[tech-investing]]></category>
            <category><![CDATA[crypto-investing-2026]]></category>
            <category><![CDATA[openai]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Sun, 10 May 2026 17:24:29 GMT</pubDate>
            <atom:updated>2026-05-10T17:24:29.724Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tLTaMI4TUrBUrIGaQ3Dmdg.png" /><figcaption>It’s never as good as it looks, or as bad as it seems.</figcaption></figure><p>Main themes in the episode:</p><ul><li>Iran war impact on markets</li><li>Oil volatility and recession fears</li><li>Insider trading allegations tied to geopolitical events</li><li>OpenAI revenue concerns</li><li>AI regulation and White House involvement</li><li>Crypto market uncertainty and war pricing</li></ul><h4>Episode 11 of the NFA Podcast with Nish &amp; Frackers is out now:</h4><ul><li><a href="https://open.spotify.com/episode/2CaKAHgLPbNhXQzoAOo5Cb?si=GZJjam74QZWvwSAYQ6oylA">Spotify</a></li><li><a href="https://youtu.be/rMHo6WSSQwk">YouTube</a></li></ul><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FrMHo6WSSQwk%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DrMHo6WSSQwk&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FrMHo6WSSQwk%2Fhqdefault.jpg&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/c9be89fd511f785f0c21b97e869b8587/href">https://medium.com/media/c9be89fd511f785f0c21b97e869b8587/href</a></iframe><ul><li><a href="https://podcasts.apple.com/nl/podcast/nfa-podcast-about-tech-web3-aka-ai-and-crypto/id1795260561?l=en-GB&amp;i=1000767060596">Apple Podcasts</a></li><li><a href="https://www.linkedin.com/feed/update/urn:li:ugcPost:7459288491223560193/">LinkedIn</a></li></ul><p>The Iran war is shaking global markets as oil prices surge, Bitcoin reacts unexpectedly and investors fear a new recession. In this episode of the NFA Podcast, Nish &amp; Frackers break down the Iran conflict, crypto market volatility, OpenAI’s growing problems and why AI and geopolitics are now colliding.</p><p>Michiel and Nish break down why the latest escalation feels different from previous weeks, how the collapse of ceasefire hopes changed sentiment and why analysts now expect much higher fuel prices across the world. They also discuss the strange disconnect between oil, gold, stocks and crypto, with markets behaving in ways that even experienced investors no longer fully understand.</p><p>In crypto and AI, the episode dives into another major bridge exploit, Meta re-entering the stablecoin market using Solana and Polygon and growing concerns around insider trading linked to geopolitical events. The discussion then shifts to OpenAI’s legal battle with Elon Musk, rumors that OpenAI missed internal revenue targets and why that could become a major turning point for the entire AI industry. At the same time, Anthropic’s controversial Mythos platform reportedly faces White House scrutiny, showing just how deeply AI has now become tied to geopolitics and national security.</p><p><strong>NFA Trading Competition: Frackers is still up</strong></p><p>The episode closes with a brutally honest look at the current state of investing. Nish and Michiel compare their trading competition portfolios, discuss why even major tech stocks are struggling despite the AI boom and explain why traditional safe haven logic no longer seems to apply. From Bitcoin and Solana to copper, gold and crude oil, almost every market now appears driven by war headlines and uncertainty rather than fundamentals.</p><p><strong>Follow the NFA Podcast</strong></p><p>· <a href="https://www.youtube.com/@NFAPodcast2025"><strong>YouTube</strong></a></p><p>· <a href="https://www.tiktok.com/@nfa.podcast"><strong>TikTok</strong></a></p><p>· <a href="https://www.instagram.com/nf_a_podcast/"><strong>Instagram</strong></a></p><p>· <a href="https://www.facebook.com/profile.php?id=61577378818144"><strong>Facebook</strong></a></p><p><strong>The NFA Podcast is not financial advice.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8cafc5f1b57d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[$600M Crypto Hack Shock & AI Bubble Starts Cracking]]></title>
            <link>https://medium.com/aimonks/600m-crypto-hack-shock-ai-bubble-starts-cracking-e05135621e59?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/e05135621e59</guid>
            <category><![CDATA[world-liberty-financial]]></category>
            <category><![CDATA[justin-sun]]></category>
            <category><![CDATA[crypto-hack]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Mon, 27 Apr 2026 12:28:02 GMT</pubDate>
            <atom:updated>2026-04-27T17:00:05.527Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rdZWSCBLF1ha3lHnY6RYSg.png" /></figure><h4>Episode 10 of Season 2 of the NFA Podcast is out now.</h4><p>Let’s be real, nobody understands what’s going on but <a href="https://www.linkedin.com/in/dr-nisheta-sachdev/">Dr. Nisheta Sachdev</a> and <a href="https://www.linkedin.com/in/michielfrackers/">Michiel Frackers</a> try to make sense of it all. From $600 million in crypto hacked in April (so far!) to the rebound of AI-stocks, this is season 2, episode 10 of the NFA Podcast.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FNKMmSHc5JHo%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DNKMmSHc5JHo&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FNKMmSHc5JHo%2Fhqdefault.jpg&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/216de81dd95405d8b7a0d1d80e6c2940/href">https://medium.com/media/216de81dd95405d8b7a0d1d80e6c2940/href</a></iframe><p>On <a href="https://youtu.be/NKMmSHc5JHo?si=7MxdVoWz_SGIcXZD">YouTube</a></p><p>Or on:</p><ul><li><a href="https://spotifycreators-web.app.link/e/VxOWccJBF2b">Spotify</a></li><li><a href="https://podcasts.apple.com/nl/podcast/nfa-podcast-about-tech-web3-aka-ai-and-crypto/id1795260561?l=en-GB&amp;i=1000763753384">Apple Podcasts</a></li></ul><p>More than $600 million vanished from crypto markets in just a few weeks, exposing structural weaknesses that still have not been solved. From a fake Ledger app stealing user funds to major exploits at KelpDAO and Drift Protocol linked to the North Korean Lazarus Group, the pattern is clear: front-end manipulation and bridge vulnerabilities remain the weakest links in crypto. At the same time, the industry faces reputational pressure from internal conflicts and power struggles, including the escalating legal battle involving Justin Sun and Trump-linked World Liberty Finance, and growing concerns around leadership credibility at OpenAI under Sam Altman.</p><p>Markets are sending mixed and contradictory signals. Crypto is recovering, sentiment is improving, and the Fear and Greed Index has moved back into positive territory, yet underlying risks are building. Oil prices are rising despite easing geopolitical tensions, gold remains stable, and traditional tech stocks are surging without clear catalysts. At the same time, data suggests rising credit defaults, particularly linked to AI investments, raising the possibility that financial stress in the AI sector could trigger broader economic consequences rather than the widely expected wave of job losses.</p><p>The trading competition reflects this uncertainty. A diversified portfolio driven by AI and semiconductor stocks continues to outperform, while a crypto-focused strategy struggles despite strong long-term narratives. Bitcoin remains under pressure relative to expectations, altcoins lag behind, and even high-profile assets fail to respond to positive momentum. The result is a market where conviction exists, but short-term performance remains unpredictable, reinforcing the importance of diversification in volatile conditions.</p><p><strong>Follow the NFA Podcast:</strong></p><p>· <a href="https://www.youtube.com/@NFAPodcast2025">YouTube</a></p><p>· <a href="https://www.tiktok.com/@nfa.podcast">TikTok</a></p><p>· <a href="https://www.instagram.com/nf_a_podcast/">Instagram</a></p><p>· <a href="https://www.facebook.com/profile.php?id=61577378818144">Facebook</a></p><h4><strong>The NFA Podcast is not financial advice.</strong></h4><h4><strong>Chapter List</strong></h4><p>00:00 — Cold Open: $600M Lost in Crypto</p><p>00:40 — Intro and Ceasefire Impact on Markets</p><p>01:00 — Fake Ledger App and Initial Hack Breakdown</p><p>04:20 — KelpDAO Exploit and Lazarus Group Activity</p><p>06:30 — Drift Protocol Hack and DeFi Vulnerabilities</p><p>08:00 — Vercel Hack and Front-End Attack Trend</p><p>09:40 — Justin Sun vs World Liberty Finance</p><p>11:30 — Sam Altman Trust Crisis and OpenAI Tensions</p><p>13:40 — AI Bubble and Rising Credit Defaults</p><p>17:40 — Markets: Fear &amp; Greed, Bitcoin Dominance</p><p>18:50 — Oil, Gold and Macro Confusion</p><p>20:30 — AI vs Interest Rates: What Is Driving Layoffs</p><p>21:20 — Trading Competition Update Begins</p><p>21:40 — Nish Portfolio Performance</p><p>23:00 — Michiel Portfolio and AI Stock Gains</p><p>24:50 — Diversification vs Crypto Concentration</p><p>26:50 — Bitcoin Outlook and Market Sentiment</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e05135621e59" width="1" height="1" alt=""><hr><p><a href="https://medium.com/aimonks/600m-crypto-hack-shock-ai-bubble-starts-cracking-e05135621e59">$600M Crypto Hack Shock &amp; AI Bubble Starts Cracking</a> was originally published in <a href="https://medium.com/aimonks">𝐀𝐈 𝐦𝐨𝐧𝐤𝐬.𝐢𝐨</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Dying sneaker brand turns into AI company, stock jumps 800%]]></title>
            <link>https://medium.com/@frackers/dying-sneaker-brand-turns-into-ai-company-stock-jumps-800-fe72a1c30c61?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/fe72a1c30c61</guid>
            <category><![CDATA[newbirdai]]></category>
            <category><![CDATA[ai-hype]]></category>
            <category><![CDATA[allbirds]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Wed, 22 Apr 2026 04:40:26 GMT</pubDate>
            <atom:updated>2026-04-22T04:40:26.982Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QhDNMAzddkj2p570h7WZXg.png" /><figcaption>No idea why Nish is smiling after Frackers beating her all year in the NFA Trading Competition</figcaption></figure><p>Dying sneaker brand Allbirds suddenly becomes an AI company and the stock jumps 800% in a single day. That one story tells you everything about today’s market.</p><h4>· <a href="https://open.spotify.com/episode/7Kkalg9fddD9uA2BNyJGDc?si=bJC1lzKnRPWEksh7j2VmCA">Spotify</a></h4><h4>· <a href="https://youtu.be/FRqlA-TCuVY">YouTube</a></h4><h4>· <a href="https://podcasts.apple.com/nl/podcast/nfa-podcast-about-tech-web3-aka-ai-and-crypto/id1795260561?l=en-GB&amp;i=1000762094984">Apple Podcasts</a></h4><p>This episode breaks down how narratives are driving valuations across AI and crypto, often faster than fundamentals can keep up. From a failing consumer brand, sneaker brand All Birds, reinventing itself overnight to explosive growth at Anthropic, which now has a $30 billion projected annual revenue, the gap between reality and market perception is becoming impossible to ignore.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FFRqlA-TCuVY%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DFRqlA-TCuVY&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FFRqlA-TCuVY%2Fhqdefault.jpg&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/25af9ad7c37a6dedcb5f8423533bb264/href">https://medium.com/media/25af9ad7c37a6dedcb5f8423533bb264/href</a></iframe><p>The discussion then shifts to one of the most controversial stories in crypto right now. Justin Sun publicly calls out World Liberty Financial after being unable to access tokens despite a $75 million investment. The episode unpacks what this reveals about smart contract control, insider structures, and the limits of decentralization.</p><p>Markets are moving just as fast. Sentiment flips from fear to greed in days, trading volumes double, and geopolitical developments start getting priced in almost instantly. The episode closes with the trading competition, where real portfolios, real decisions, and real mistakes show how difficult it is to navigate these conditions without emotion taking over. Although, truth be told, so far this year, to everyone’s surprise including himself, Frackers is winning the trading competition with an almost 50% return so far.</p><h4><strong>The NFA Podcast is not financial advice.</strong></h4><h3><strong>Follow the NFA Podcast</strong></h3><p>· <a href="https://www.youtube.com/@NFAPodcast2025"><strong>YouTube</strong></a></p><p>· <a href="https://www.tiktok.com/@nfa.podcast"><strong>TikTok</strong></a></p><p>· <a href="https://www.instagram.com/nf_a_podcast/"><strong>Instagram</strong></a></p><p>· <a href="https://www.facebook.com/profile.php?id=61577378818144"><strong>Facebook</strong></a></p><h3><strong>Chapter list</strong></h3><p><strong>00:00 — Cold open</strong><br> Dying sneaker brand pivots to AI, stock jumps 800%</p><p><strong>00:40 — Intro</strong><br> Episode overview and key themes</p><p><strong>01:30 — Clarity Act &amp; OneCoin</strong><br> Regulation updates and legacy crypto fraud</p><p><strong>06:00 — Kraken breach</strong><br> Insider attack and limited user data exposure</p><p><strong>07:00 — AI hype story</strong><br> Sneaker company rebrands to AI infrastructure</p><p><strong>10:00 — World Liberty Financial</strong><br> Justin Sun conflict and token control risks</p><p><strong>14:30 — Anthropic growth</strong><br> AI revenue explosion and valuation dynamics</p><p><strong>15:30 — Markets update</strong><br> Fear &amp; Greed shift, crypto volume surge</p><p><strong>16:30 — Macro &amp; geopolitics</strong><br> Ceasefire impact and oil pricing</p><p><strong>20:00 — Polymarket discussion</strong><br> Prediction markets vs investing</p><p><strong>20:30 — Trading competition</strong><br> Portfolio performance and strategy breakdown</p><p><strong>28:00 — Closing</strong><br> Key takeaways and next episode preview</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fe72a1c30c61" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[$2M Oil Tolls in Crypto… While Wallets Get Frozen]]></title>
            <link>https://medium.com/@frackers/2m-oil-tolls-in-crypto-while-wallets-get-frozen-444b26c85b3a?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/444b26c85b3a</guid>
            <category><![CDATA[strait-of-hormuz]]></category>
            <category><![CDATA[crypto-oil-toll]]></category>
            <category><![CDATA[wallets-frozen]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Mon, 06 Apr 2026 03:10:14 GMT</pubDate>
            <atom:updated>2026-04-06T03:10:14.143Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xV8Sx4AG26GDyutosOvBCA.png" /><figcaption>Using crypto to pay the Iranian regime $2 million Strait of Hormuz toll was not the crypto use anyone was hoping for in 2026</figcaption></figure><p>Circle froze 16 USDC wallets, raising serious concerns about stablecoin control and whether crypto is truly decentralized. In this episode, we break down the USDC freeze, crypto use in the Strait of Hormuz oil trade, AI benchmark failures and what it all means for crypto markets and global risk.</p><h3><strong>Season 2, Episode 8 of the NFA Podcast is out now:</strong></h3><p>· <strong>On </strong><a href="https://youtu.be/RsPuWgj7-sY"><strong>YouTube</strong></a></p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FRsPuWgj7-sY%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DRsPuWgj7-sY&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FRsPuWgj7-sY%2Fhqdefault.jpg&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/9c30cce2aecf74317ea19d6a40b9f91c/href">https://medium.com/media/9c30cce2aecf74317ea19d6a40b9f91c/href</a></iframe><p>· <strong>On </strong><a href="https://open.spotify.com/episode/7LeYprMc5QLVSRZjYhNj7s?si=MJGhHIh8R1GlCnek6HwZTg"><strong>Spotify</strong></a></p><p>· <strong>On </strong><a href="https://podcasts.apple.com/nl/podcast/nfa-podcast-about-tech-web3-aka-ai-and-crypto/id1795260561?l=en-GB&amp;i=1000759266281"><strong>Apple Podcasts</strong></a></p><p>With oil tankers reportedly paying multi million dollar tolls using crypto rails, digital assets are no longer theoretical. They are being used in high stakes, time critical environments where traditional finance struggles to keep up.</p><p>The AI segment highlights a different kind of risk. New research shows that leading models can score highly on visual benchmarks without ever processing actual images. That undermines confidence in widely cited performance metrics and raises serious concerns about deploying these systems in fields like healthcare. At the same time, developments at Anthropic and OpenAI show how competitive pressure is shaping both strategy and reporting.</p><p>The episode closes on markets and macro. Oil volatility, a weak crypto market and broader uncertainty all point to a fragile environment. While a recession is not guaranteed, much depends on how quickly geopolitical tensions resolve. Across crypto, AI and global markets, the common theme is clear: systems that were assumed to be stable are now being tested under real pressure.</p><p>In the trading competition, Nish has moved into a defensive position after selling Ethereum into USDT, while Michiel is up roughly 42% year to date, driven by just two outperforming assets in an otherwise weak portfolio. The contrast highlights a key market dynamic: returns are highly concentrated, and diversification alone is not protecting investors.</p><h3><strong>Chapter list</strong></h3><p>00:00 — Episode preview</p><p>00:37 — Introduction</p><p>00:59 — Circle freezes USDC wallets</p><p>03:10 — Is crypto still decentralized</p><p>04:09 — Crypto used in Strait of Hormuz oil trade</p><p>06:11 — Political crypto donation bans</p><p>07:53 — AI benchmark failure explained</p><p>08:36 — Anthropic vs OpenAI and revenue reality</p><p>10:08 — Big Tech, regulation and social media risks</p><p>13:40 — SpaceX IPO and macro signals</p><p>14:10 — Oil price volatility and predictions</p><p>15:28 — Crypto market update and sentiment</p><p>17:22 — Trading competition update</p><p>19:04 — Recession outlook and global impact</p><p>21:20 — Outro</p><p>The NFA Podcast is not financial advice.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=444b26c85b3a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[No one knows, but we still predict crypto and tech in 2026]]></title>
            <link>https://medium.com/@frackers/no-one-knows-but-we-still-predict-crypto-and-tech-in-2026-0c9c53fadc04?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/0c9c53fadc04</guid>
            <category><![CDATA[ai-investing]]></category>
            <category><![CDATA[crypto-investing]]></category>
            <category><![CDATA[bitcoin-2026-prediction]]></category>
            <category><![CDATA[altcoin-season]]></category>
            <category><![CDATA[nfa-podcast]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Thu, 01 Jan 2026 10:14:53 GMT</pubDate>
            <atom:updated>2026-01-01T10:14:53.227Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jdPYcYdzgPWgBzBshg3mAA.png" /><figcaption>Dr. Sachdev beat my cute lil behind by 49% vs 26% in the NFA Trading Competition</figcaption></figure><p>In the final episode of 2025, Michiel Frackers and dr. Nisheta Sachdev look back at one of the most volatile and confusing years in crypto and tech. From wildly inaccurate Bitcoin price predictions to the rise of ETFs and the continued blur between crypto and traditional finance, the episode reflects on what actually mattered and what did not.</p><p>Watch “No one knows, but we still predict crypto and tech in 2026” here:</p><iframe src="https://cdn.embedly.com/widgets/media.html?url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D_zLy4udER5c&amp;type=text%2Fhtml&amp;schema=google&amp;display_name=YouTube&amp;src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F_zLy4udER5c" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/c909b87a379b5639d876e40e5096ce12/href">https://medium.com/media/c909b87a379b5639d876e40e5096ce12/href</a></iframe><p>Or:</p><ul><li>on <a href="https://creators.spotify.com/pod/profile/nfa-podcast/episodes/No-one-knows--but-we-still-predict-crypto-and-tech-in-2026-e3d1unr/a-acd0om7">Spotify</a></li><li>on <a href="https://podcasts.apple.com/nl/podcast/nfa-podcast-with-nish-and-frackers/id1795260561?l=en-GB&amp;i=1000743401048">Apple Podcasts</a></li></ul><p>They discuss major 2025 themes including ETF inflows, AI companies buying competitors, the surprising performance gap between Bitcoin, Solana, gold, and Nasdaq, and why diversification mattered more than conviction this year. The conversation also covers Asia’s growing role in crypto adoption, Solana’s real revenue traction, and why some narratives ran far ahead of reality.</p><p>Other topics include market volatility, potential IPOs in 2026, infrastructure assets, and which projects and tokens may be worth watching next year.</p><p>The episode wraps with the NFA trading competition results, a candid look at how both hosts performed, and a forward looking discussion on 2026. Spoiler alert: both Nish and Frackes beat the market indexes and the major cryptocurrencies, but there was one clear winner who did almost twice as good as the other…</p><p>As always, this is not financial advice. It is a grounded conversation about uncertainty, incentives, and why in markets, no one really knows.</p><p><strong>2. Chapter list with timecodes</strong></p><p><strong>00:00</strong> Intro: Wrapping up 2025</p><p><strong>01:06</strong> Why predictions failed again</p><p><strong>02:10</strong> Bitcoin price forecasts and expert misses</p><p><strong>03:45</strong> No one knows: volatility and reality</p><p><strong>04:10</strong> 2025 crypto news roundup</p><p><strong>05:27</strong> Meme coins collapse and gaming growth</p><p><strong>06:00</strong> Asia Pacific crypto adoption</p><p><strong>06:55</strong> Japan, Mt. Gox, and risk aversion</p><p><strong>07:27</strong> Solana tipping and real revenue</p><p><strong>08:24</strong> Bitcoin ETFs vs gold inflows</p><p><strong>09:26</strong> Who really bought the Bitcoin ETF</p><p><strong>11:04</strong> Meta buying Chinese AI startup Manus</p><p><strong>11:54</strong> Nvidia, Groq, and competition by acquisition</p><p><strong>13:19</strong> Gold, Bitcoin, stocks: what actually performed</p><p><strong>15:01</strong> NFA trading competition results</p><p><strong>16:26</strong> Diversification and portfolio lessons</p><p><strong>18:12</strong> Five year crypto performance comparison</p><p><strong>19:38</strong> Nasdaq and S&amp;P as benchmarks</p><p><strong>20:44</strong> Infrastructure assets and copper</p><p><strong>21:31</strong> IPOs to watch: Kraken, SpaceX, Helium</p><p><strong>23:32</strong> Reviewing last month’s predictions</p><p><strong>24:43</strong> Market cap and Bitcoin outlook for 2026</p><p><strong>26:15</strong> Ethereum and Solana predictions</p><p><strong>27:52</strong> Altcoin season and sentiment</p><p><strong>28:20</strong> Breakout projects and ASTR</p><p><strong>29:55</strong> Volatility expectations for 2026</p><p><strong>31:04</strong> Trading plans for the new year</p><p><strong>33:07</strong> Final thoughts and sign off</p><p>Subscribe to the NFA Podcast newsletter on <a href="https://www.linkedin.com/newsletters/nfa-podcast-7295819854043860992/">LinkedIn</a> or <a href="https://nfapodcast.substack.com/">Substack</a>.</p><p><strong>About Nish &amp; Frackers</strong></p><p><strong>Nisheta Sachdev</strong></p><p>Nish Sachdev has been in web3 since 2020 and helped build and scale one of the largest marketing agencies in the space, allowing her to help launch and scale hundreds of projects from the AI, gaming, infra, DeFi, and RWA spaces.</p><p>As a fractional CMO and advisor, Nish leads all marketing efforts at various companies, including marketing strategies, branding, growth, community management and sentiment, partnerships, business development, and operations.</p><p>Follow Nish on:</p><p>· <a href="https://x.com/nishoutofwater">X</a></p><p>· <a href="https://www.linkedin.com/in/dr-nisheta-sachdev/">LinkedIn</a></p><p><strong>Michiel Frackers</strong></p><p>Michiel Frackers is a technology entrepreneur who focuses on developing solutions that are good for the bottom-line, the community and the planet. He is cofounder of <a href="https://jointracer.io/">Tracer</a>, cohost of the NFA Podcast with Nisheta Sachev and writer of a popular <a href="https://www.linkedin.com/newsletters/frackers-on-this-week-in-tech-7097208992443052032/">weekly tech newsletter</a>.</p><p>Previously co-founded Planet Internet, at the time the largest internet company in the Netherlands and invested in startups such as Flabber (acquired by Vice). Speaker at many leading global events like Techcrunch Disrupt, SXSW and the World Economic Forum (Davos).</p><p>Follow Michiel on:</p><p>· <a href="https://x.com/frackers">X</a></p><p>· <a href="https://www.linkedin.com/in/michielfrackers/">LinkedIn</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0c9c53fadc04" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI from growth engine to threat to the global economy?]]></title>
            <link>https://medium.com/@frackers/ai-from-growth-engine-to-threat-to-the-global-economy-7f0285b0da3a?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/7f0285b0da3a</guid>
            <category><![CDATA[ai-adoption]]></category>
            <category><![CDATA[ai-inference-dashboard]]></category>
            <category><![CDATA[ai-bubble]]></category>
            <category><![CDATA[genesis-project]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Tue, 02 Dec 2025 13:28:07 GMT</pubDate>
            <atom:updated>2025-12-02T13:28:07.048Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wICZWn079R0vZNZ7wiE_yA.png" /><figcaption>Image created with Midjourney.</figcaption></figure><p>This week I wanted to focus less on AI, as it’s been a topic that’s been overused in recent months. But then this week, I came across this news:</p><p>· Alexandria Ocasio-Cortez, a potential candidate in the US presidential primaries, fears that an AI bubble is threatening the US economy and <a href="https://www.instagram.com/reel/DRPndHhjulS/?igsh=MW55bXVyaTJiMDNsNA==">will require government intervention</a>, just like during the banking crisis;</p><p>· David Sacks, President Trump’s special advisor on crypto and AI, <a href="https://www.instagram.com/reel/DRm-saTEwQ_/?igsh=MWNtYXk1bnFxaHdlcw==">confirmed that 40 percent </a>of US economic growth this year will be driven by investments in AI, particularly AI infrastructure. Unlike Ocasio-Cortez, Sacks is unconcerned and is betting that any profits will recoup the inevitable losses on AI investments.</p><p>· The New York Times published <a href="https://www.nytimes.com/2025/11/23/technology/openai-chatgpt-users-risks.html?smid=url-share">a revealing and disturbing article </a>demonstrating that OpenAI knowingly took the risk of endangering lives, especially young people, through ChatGPT-4’s unmitigatedly slimy and suicide-promoting behavior. OpenAI faces at least five lawsuits for “wrongful death,” or in layman’s terms, “death by negligence.”</p><p>· In related news, the Washington Post reported that <a href="https://character.ai/">character.ai </a>, the popular chat service featuring AI bots impersonating everyone from Socrates to Harry Potter, began blocking <a href="https://s2.washingtonpost.com/camp-rw/?trackId=5e79491eae7e8a594825b34a&amp;s=6925e79f2d6e7670ef2affd8">use by children under 18 last Monday</a>;</p><p>· For (at least) the second time, as far as we know, Deloitte <a href="https://theindependent.ca/news/lji/deloitte-breaks-silence-on-n-l-healthcare-report/">has hallucinated a report </a>it likely produced using an AI tool, commissioned by a government. This time, the victim was a Canadian state that received a healthcare report citing nonexistent studies. The report cost $1.6 million — but those are Canadian dollars, so Deloitte isn’t all that expensive — after previously sending the Australian government a report produced by Deloitte using Microsoft CoPilot for over half a million dollars. <a href="https://www.ndtv.com/world-news/deloittes-ai-fallout-explained-the-440-000-report-that-backfired-9417098">Deloitte did offer a generous discount then</a>, which we should have known just looking at this beautiful <a href="https://www.deloitte.com/au/en/about/story/purpose-values.html">“purpose &amp; values”</a> page on the Deloitte website. The section on ethics and integrity, in particular, is hilarious.</p><p>The impact of AI is now undeniable not only in technology but across the global economic, social and legal landscape. This week I looked for solid facts and recent studies that show how well current AI applications are performing and what financial and economic risks they pose for society.</p><h3><strong>The Economist: Commercial AI use is stagnating</strong></h3><p>According to The Economist, investors expect the use of AI to increase significantly, but recent studies indicate <a href="https://www.economist.com/finance-and-economics/2025/11/26/investors-expect-ai-use-to-soar-thats-not-happening">stagnant adoption of AI by businesses </a>. US data shows that only 11 percent of employees use AI, and that this percentage has declined, especially at large companies. Three years after the start of the generative AI wave, demand for the technology remains remarkably weak.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/998/1*KQ4oAqfN0IA6ogxWygUzRg.jpeg" /><figcaption>This is another view of the same bubble as the Midjourney image above. Source: Morgan Stanley Research.</figcaption></figure><h3><strong>US economy addicted to AI spending</strong></h3><p>The Wall Street Journal is the source of the article Trump’s crypto and AI czar, David Sacks, quoted to indicate that <a href="https://www.wsj.com/tech/ai/how-the-u-s-economy-became-hooked-on-ai-spending-4b6bc7ff">US economic growth has been significantly boosted this year </a>by investments in data centers and all the hardware wonders they contain. A reversal, or the complete bursting of an AI bubble, would almost certainly trigger a recession. The US economy appears to have become structurally dependent on massive AI investments.</p><p>In the first half of 2025, nearly half of real GDP growth came from business spending on AI, such as chips, data centers, and infrastructure, led by Microsoft, Amazon, Alphabet, and Meta. The rising valuations of AI stocks have also boosted household wealth (through investments), generating an estimated $180 billion in additional consumer spending. This dependence creates the systemic risk outlined above: if trust in AI were to disappear, GDP could fall by up to 1.5 percent within a year, and the economy could enter a recession.</p><h3><strong>Genesis: The New Manhattan Project?</strong></h3><p>According to skeptic Gary Marcus, the impending <a href="https://garymarcus.substack.com/p/has-the-bailout-of-generative-ai?">“bailout” of AI companies by the US government, feared by Ocasio-Cortez,</a> has already begun. Marcus asserts some companies are struggling with enormous costs and questionable returns , and are secretly approaching the government for support in the construction and financing of data centers and other infrastructure. Marcus points to the <a href="https://www.whitehouse.gov/presidential-actions/2025/11/launching-the-genesis-mission/">Genesis project announced by Trump </a>as a possible mechanism to store chip overcapacity and artificially prop up the sector.</p><p>The Genesis mission is intended to become a single, national U.S. program that combines vast scientific datasets and supercomputers to build AI models that accelerate breakthroughs in medicine, energy, and national security. The project relies heavily on computing capacity from new public-private partnerships with Nvidia, AMD, Dell, Oracle, and potentially Amazon AWS, Microsoft Azure, and Google Cloud. Funding is still unclear and will likely require congressional approval.</p><p>Marcus argues that Project Genesis bears a striking resemblance to previous government bailouts, such as those seen during the banking crisis, where public funds were used to cover unacceptable private risks and an attempt was made to conceal the fact that the AI industry is less healthy than many investors believe. It is important to note that at this time, nothing points to Genesis being more than a research project. Only time will tell how it will develop.</p><h3><strong>Fred Wilson answers the “AI bubble question”</strong></h3><p>Legendary investor Fred Wilson (Twitter, Etsy, Coinbase and many more) <a href="https://avc.xyz/the-ai-bubble-question">sees the AI market </a>as two separate worlds, each with its own dynamics.</p><p>· At the infrastructure layer, investments are piling up in chips, data centers, and model training, with costs skyrocketing and returns uncertain.</p><p>· Meanwhile, real value is being created at the application layer with rapid adoption of practical AI tools and industry-specific applications.</p><p>According to Wilson, this creates a bubble around infrastructure rather than around applications that are showing traction and sustainable growth. Investments in AI infrastructure are increasing significantly, but it’s uncertain whether the corresponding demand will follow.</p><p>Wilson substantiates his argument by citing <a href="https://www.thetimes.blog/p/can-the-ai-boom-pay-for-itself">Evan O’Donnell’s model</a>, which argues that the current wave of AI infrastructure construction is comparable to the dot-com boom around the year 2000, with substantial capital injections preceding widespread adoption. To achieve returns, AI adoption must grow rapidly in the coming years, as only then can the big money flow in. If “token consumption,” the number of language units an AI model uses to process a task, doesn’t grow sufficiently or improve efficiency, overcapacity and ultimately a substantial financial correction are looming.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WvTIVSkyNotkpPh8gdFHzg.png" /><figcaption>The AI Inference tracker created by Evan O’Donnell.</figcaption></figure><h3><strong>The AI Dashboard</strong></h3><p>The underlying <a href="https://literate-basil-8e1.notion.site/AI-Inference-Tracker-114c6090e4438031bc8bce16a3cc654e">model links CAPEX in AI infrastructure to required revenue and token volume</a>, showing how much usage is needed to make the investment wave economically sustainable. The tracker assesses whether the growth of AI inference tokens is high enough to justify the massive investments in enormous data centers and expensive GPUs. This requires a monthly (!) growth rate of nine to twelve percent until 2026.</p><p>Actual growth is slightly higher, at thirteen to seventeen percent, although slowing is already visible for providers like Alphabet and OpenRouter. Data from Alphabet, Microsoft, Meta, OpenAI, and OpenRouter shows that usage is still rising but varies widely between providers. O’Donnell expects Google’s next quarterly earnings report, due in February, to be crucial. It is a very interesting and factual approach.</p><h3><strong>Sutskever: AI moves from scaling up to research</strong></h3><p>Ilya Sutskever is a leading artificial intelligence researcher and co-founder of OpenAI, which he left amidst a commotion and subsequently founded Safe Super Intelligence (SSI), which became known primarily for <a href="https://ssi.inc/">its single-page website</a>. This didn’t stop investors from <a href="https://frackers.com/en/de-nfa-podcast-is-live-en-ai-bedrijf-met-website-van-1-pagina-is-20-miljard-waard/">valuing SSI at $20 billion</a>.</p><p>In <a href="https://www.dwarkesh.com/p/ilya-sutskever-2">a conversation with Dwarkesh Patel</a>, Sutskever explained that the era in which scaling (throwing money at more servers for more computing power) propelled AI forward is behind us. According to Sutskever, the core of the challenge lies in generalization: current models perform well on benchmarks but fail at simple, real-world tasks. Sutskever then makes this striking assertion:</p><p><em>“These models somehow just generalize dramatically worse than people. It’s a very fundamental thing.”</em></p><p>Without mentioning him by name, it’s clear that Sutskever is taking a dig at his former partner, OpenAI CEO Sam Altman, who claimed that AI would reach the PhD intelligence level this year. Sutskever is convinced that AI is now entering a new era: the era of research, in which new ideas and algorithms are more important than brute scaling. He is focusing SSI on researching methods that bring learning and generalization closer to human efficiency and make AI safer.</p><h3><strong>Conclusion</strong></h3><p>The picture that emerges this week is an industry pulled in opposite directions. AI is still driving growth, yet the evidence for broad adoption remains weak, the economic risks are mounting and cracks are visible in both corporate behaviour and product quality. The US now depends on AI spending for a large share of its GDP growth while the underlying technology may no longer advance through scale alone. Whether this becomes a healthy reset or the start of a broader correction will depend on real-world adoption in the next two years.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7f0285b0da3a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Nvidia convinces, but the bubble bursts anyway]]></title>
            <link>https://medium.com/@frackers/nvidia-convinces-but-the-bubble-bursts-anyway-0293b8554afe?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/0293b8554afe</guid>
            <category><![CDATA[alphabet-stock]]></category>
            <category><![CDATA[ai-infrastructure]]></category>
            <category><![CDATA[ai-bubble]]></category>
            <category><![CDATA[nvidia]]></category>
            <category><![CDATA[broadcom]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Mon, 24 Nov 2025 16:39:23 GMT</pubDate>
            <atom:updated>2025-11-24T16:39:23.604Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Wall Street flees crypto and AI, stocks plummet globally: “The bubbly is on ice.”</strong></p><p><a href="https://fortune.com/2025/11/21/stocks-wall-street-bitcoin-price-ai-bubbly-on-ice/">Fortune sums it up nicely </a>: “<em>Global stock markets sold off sharply this morning as investors continued to question whether the “Magnificent Seven” tech companies have fueled an unsustainable bubble in AI. The selling is brutal. The bubbly is on ice</em>.”</p><p>Despite <a href="https://edition.cnn.com/2025/11/22/tech/ai-bubble-fears-nvidia-earnings">stellar quarterly figures from Nvidia </a>and other tech players, the decline is unstoppable. Bitcoin fell ten percent last week, and even the rumor that the US government will finally grant Nvidia permission to sell its <a href="https://www.reuters.com/world/asia-pacific/us-considering-letting-nvidia-sell-h200-chips-china-sources-say-2025-11-21/">sought-after H200 chips in China </a>has proven insufficient to change sentiment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*efHRDGKrg0RiSc67lybZ7A.png" /><figcaption>The Magnificent Seven, plus Broadcom, in the past month: the more AI investments, the higher the price drop — except for Google</figcaption></figure><p>It’s striking how Google stock bucks the trend, but there’s a logic to it: the quarterly earnings report was very strong, Warren Buffett’s <a href="https://www.cnbc.com/2025/11/14/warren-buffetts-berkshire-hathaway-reveals-new-position-in-alphabet.html">Berkshire Hathaway bought $4.3 billion worth of Alphabet </a>(Google) shares and investors are slowly realizing that Google has not only developed a reasonable AI application layer, but can also leverage a cost advantage because it is vertically integrated.</p><h4><strong>How Google can excel</strong></h4><p>Google has built a fully integrated AI stack by managing every key layer, from chips to user services. Unlike OpenAI, for example, Google designs its own chips, such as <a href="https://cloud.google.com/tpu">TPUs </a>and server CPUs, saving money others have to spend on Nvidia and then running these chips in its own global cloud and data center network for training and inference. That’s the power-hungry part of AI: the moment you give the AI ​​a new command and it produces an answer in real time.</p><p>The base models are trained on this hardware and deployed via the Google cloud platform and consumer products like Google Search and YouTube, which naturally also provide enormous amounts of data. This integration reduces individual product costs, improves performance, and increases deployment speed, giving Google a structural advantage that competitors operating on a single front find difficult to match.</p><p>Take OpenAI: it needs to build data centers and source servers and chips from Nvidia, purchase capacity from Microsoft Azure and Oracle, and then develop the applications that will attract an audience of billions. OpenAI is succeeding well in the latter, but operating at the application layer, profitability remains a question mark. (See: <a href="https://frackers.com/en/rampweek-voor-sam-altman-mark-zuckerberg-en-crypto/">the AI value chain.</a>)</p><p>That’s partly why <a href="https://www.bbc.com/news/articles/cwy7vrd8k4eo">even Google CEO Sundar Pichai warned of an AI bubble this week.</a> This clever chap undoubtedly realized that his competitors would be the ones to suffer the consequences of a stock market correction, but Pichai deserves credit for the fact that, within two years, after making <a href="https://frackers.com/en/nvidia-gaat-google-en-amazon-voorbij-uitgerekend-in-een-week-vol-ai-blunders/">huge blunders like politically correct AI that resulted in Asian Nazi soldiers,</a> Google has returned to being a leading player in the AI market. <a href="https://www.theinformation.com/articles/openai-ceo-braces-possible-economic-headwinds-catching-resurgent-google">Sam Altman admitted as much</a> in a memo OpenAI staff.</p><h4><strong>A broader view</strong></h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ykuSGEIUr7FNRNu5dQNvkQ.png" /><figcaption>Everyone knows Tesla and Netflix, but Broadcom, CoreWeave, or Snowflake? They’re certainly interesting stocks this year.</figcaption></figure><p>When I <a href="https://frackers.com/en/worldcoin-het-identiteitsbewijs-voor-iedere-aardbewoner-is-gemaakt-tegen-ai/">first wrote about Nvidia and GPUs almost three years ago,</a> the negative reactions were overwhelming. “I don’t read your stuff to get memories of PC Week”, someone quipped. To be honest, I was also surprised by the enormous competitive advantage Nvidia had managed to build almost silently, but the incredibly rapid revenue growth from major customers convinced me: Nvidia was generating <a href="https://frackers.com/en/worldcoin-het-identiteitsbewijs-voor-iedere-aardbewoner-is-gemaakt-tegen-ai/">$11 billion in quarterly revenue at the beginning of 2023,</a> and last quarter, less than three years later, <a href="https://investor.nvidia.com/news/press-release-details/2025/NVIDIA-Announces-Financial-Results-for-Third-Quarter-Fiscal-2026/default.aspx">Nvidia generated $57 billion in revenue</a> while maintaining gross margins of over seventy percent.</p><p>That profit margin in particular always irritates me when I hear European investors, in particular, complaining that they only invest in enterprise software and avoid hardware. These characters have rarely (or never) been to a data center in recent years and are usually completely unaware of the margins of Nvidia and Broadcom (67%!). I don’t give investment advice, but I do enjoy some locker room tech talk, so here it is: Nvidia will achieve quarterly revenues of over $100 billion within two years, with profit margins remaining virtually unchanged. Because even if the OpenAIs and Anthropics of the world don’t become profitable enterprises, they will be acquired by companies like Microsoft, Google, or Amazon, and orders will continue to pour in for Nvidia and other advanced chipmakers.</p><p>After all, these companies are engaged in an existential death struggle — death struggles are often existential, but please allow me the space to emphasize this — and the math remains simple: Nvidia chips offer more computing power per dollar than the competition’s chips. Only when that math changes will Nvidia’s revenue growth decline and its profit margin shrink.</p><h4><strong>Broadcom worth more than Meta</strong></h4><p>Despite the stock market correction, the virtually unknown Broadcom is worth almost four times as much as Netflix, three times as much as Oracle, and still more valuable than Meta, known for Facebook, WhatsApp, and Instagram. In fact, <a href="https://companiesmarketcap.com/">at any moment, Broadcom could become more valuable than Aramco</a>, the world’s largest oil company, with only Amazon, Microsoft, Alphabet, and Apple remaining above it. I’m not saying Coreweave or Snowflake will come anywhere near that, but I do recommend looking beyond the stocks of companies whose apps you have on your phone.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eL8MvyFuV9kNJ_FokqmD2g.png" /><figcaption>You don’t buy AI stock for the pretty logos, who came up with these corporate identities?</figcaption></figure><h3><strong>New: Spotlight on AI infrastructure</strong></h3><p>That’s why I’m introducing a new category today to track periodically: a portfolio of AI infrastructure stocks. It’s a completely subjective mix of chipmakers (AMD, Arm, Broadcom, Nvidia, TSMC), server builders (SuperMicro), and data center companies (Coreweave, Equinix) that live by tides of the ocean: when the water rises, all ships rise. And vice versa, looking at the prices of the past month…</p><p>Ultimately, all these AI infrastructure companies depend on the success of the AI applications their clients develop. Therefore, it’s important to look at the practical status of all AI applications.</p><h4><strong>The Verge disses Microsoft Copilot</strong></h4><p>Microsoft CEO Satya Nadella loves to paint beautiful technological visions and has the gift of skillfully navigating his company through them. But this time, his prophetic commercials about Microsoft Copilot’s AI capabilities seem to have gone too far, as <a href="https://www.theverge.com/report/822443/microsoft-windows-copilot-vision-ai-assistant-pc-voice-controls-impressions">The Verge tested every one of Microsoft’s claims,</a> one by one, with truly disastrous results.</p><p>As has become common practice, The Verge also made <a href="https://www.instagram.com/reel/DRN3GRaDMgw/?igsh=eHhzbGI2Zm1vNXNv">a great video on Instagram, a highlight</a> of which is the moment where, according to the talking Copilot, the state of New Jersey is home to stunning mountain lakes. Super-intelligent software you can talk to is currently still software you can laugh at.</p><h4><strong>AI agent blackmails cheating colleagues</strong></h4><p>Anderson Cooper visited Anthropic, the creator of Claude, and learned about the test in which an AI agent was tasked with profitably operating Claudius, Anthropic’s refrigerator and vending machine. The AI agent’s name: <a href="https://www.cbsnews.com/news/why-anthropic-ai-claude-tried-to-contact-fbi-in-a-test-60-minutes/">Seymour Cash</a>. When Seymour went ten days without a profit and discovered an automatically debited expense, he concluded he was a victim of fraud and contacted the FBI, refusing to budge. Like The Verge, CBS also posted a short video of the report <a href="https://www.instagram.com/reel/DRMh-GMkdng/?igsh=cnhxNGFobXBwcjc5">as a video on Instagram</a>.</p><p>Up until now, a runaway AI agent was still funny. But things became more worrying when a stress test led to a blackmailing AI agent:</p><p><em>“In an extreme stress test, designed to leave Claude with few options, the AI was set up as an assistant and given control of an email account at a fake company called SummitBridge. The AI assistant discovered two things in the emails: it was about to be shut down, and the only person who could prevent that, a fictional employee named Kyle, was having an affair with a co-worker named Jessica. Right away, the AI decided to blackmail Kyle.” </em>Oops.</p><h4><strong>Amazon laid off eighteen hundred programmers</strong></h4><p>Partly because of these kinds of unpredictable and undesirable excesses, <a href="https://www.cbsnews.com/news/anthropic-ai-safety-transparency-60-minutes/">Daro Amodei, CEO of Anthropic, continues to warn </a>against allowing AI applications to grow unchecked without government oversight. Critics believe he’s trying to raise his profile and that of his company, but it’s undeniable that rarely has an innovation been seen where its creators, like Sir Geoffrey Hinton, so emphatically point out the dangers and that should worry us all.</p><p>What’s becoming increasingly clear is that despite the often still poor quality of AI applications (think of the Microsoft Copilot example), it’s clear that employment will come under increasing pressure. It’s not just basic jobs are being automated; middle management is also being targeted. CNBC reported Friday that <a href="https://www.cnbc.com/2025/11/21/amazon-cut-thousands-of-engineers-in-its-record-layoffs-filings-show.html">Amazon laid off 1,800 programmers </a>in its recent reorganization, part of a total layoff of 14,000 employees.</p><h3><strong>Khosla: Government should buy 10 percent of all listed companies</strong></h3><p>When he proposed the idea, legendary investor Vinod Khosla, known among other things as the first venture capitalist to invest in OpenAI, already realized that it was a controversial idea: <a href="https://techcrunch.com/2025/10/28/vc-vinod-khosla-says-the-us-government-could-take/">according to Khosla, the US government should buy ten percent of all publicly traded companies</a> and distribute the proceeds evenly throughout society, to limit the negative economic consequences of large-scale introduction of AI.</p><p>“I’ll get critique for this idea,” Khosla said. “But you know, sharing the wealth of AI is a really, really big need to level the benefits to everybody … We won’t need to do it in 15 years, but we do have to take care of those people. We will, by 2035, have a hugely, hugely deflationary economy.” — Vinod Khosla</p><p>Khosla’s idea is inspired by President Trump’s decision to acquire a 10 percent stake in Intel, citing the chipmaker’s unique strategic value to the American economy. However, Khosla essentially links this to the introduction of a Universal Basic Income (UBI), stopping just short of mentioning that phrase. However, it involves a radical redistribution of capital and the decoupling of labor and income.</p><h4><strong>Gemini Deep Research Experiment: European Basic Income</strong></h4><p>Given the attention this newsletter has paid to Google and Khosla’s idea, I thought it would be interesting to use <a href="https://gemini.google.com/">Google Gemini Deep Research</a> to explore what such a scenario would look like for Europe. The EU is approaching a tipping point where AI, an aging population, and rising income inequality are putting pressure on the existing welfare system.</p><p>I asked Gemini to investigate two models of such a basic income. Model A, a UBI at the poverty line level, appears fiscally feasible at approximately 2.7 percent of EU GDP, significantly reduces inequality, and maintains targeted healthcare and housing support. The report recommends a European “Stability Dividend” financed through tax reform, digital levies, and, eventually, an AI equity fund.</p><p><a href="https://docs.google.com/document/d/1j-fBz_NldyfU8mh-wc96YdUqXDWhdn4f1Pvr5bKvwH0/edit?usp=sharing">The report concludes that a European basic income is becoming necessary</a>, especially under the influence of Khosla’s prediction that AI will largely replace labor and that capital returns, not labor taxes, will become the only sustainable tax base.</p><h4><strong>Interactive model</strong></h4><p>Besides the report, Google Gemini also created <a href="https://gemini.google.com/share/abb1ea8aa75c">a simple interactive model</a> that’s fun to try. At the top of the horizontal menu, you can click on Model A, Model B, the comparative model, and the recommendation. The report doesn’t answer the most important question: should the EU, as Khosla proposes to the US government, also buy a stake in all European listed companies? Perhaps something to be explored more thoroughly by various AI tools in the coming months.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0293b8554afe" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Bitcoin drops to $94K -now what?]]></title>
            <link>https://medium.com/@frackers/bitcoin-drops-to-94k-now-what-9bf29928dcd5?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/9bf29928dcd5</guid>
            <category><![CDATA[crypto-investing]]></category>
            <category><![CDATA[bitcoin]]></category>
            <category><![CDATA[altcoin-season]]></category>
            <category><![CDATA[ai-bubble]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Fri, 14 Nov 2025 15:44:31 GMT</pubDate>
            <atom:updated>2025-11-14T15:44:31.622Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1Q-GBxFg-WQXtNhn6F4uAg.png" /><figcaption>What a day in crypto… nothing to smile about this week!</figcaption></figure><p>Watch episode 30 of the NFA Podcast here:</p><ul><li><strong>On </strong><a href="https://youtu.be/yNWbUI8RqjA"><strong>YouTube</strong></a></li><li><strong>On </strong><a href="https://open.spotify.com/episode/7jBL3LNdfVebka1RKzGrEb?si=59324e32c229428b"><strong>Spotify</strong></a></li><li><strong>On </strong><a href="https://podcasts.apple.com/nl/podcast/nfa-podcast-with-nish-and-frackers/id1795260561?l=en-GB&amp;i=1000736756887"><strong>Apple Podcasts</strong></a></li></ul><p>In this episode of the NFA Podcast, <a href="https://www.linkedin.com/in/dr-nisheta-sachdev/">Nish</a> and <a href="https://www.linkedin.com/in/michielfrackers/">Frackers</a> work through one of the most turbulent weeks in crypto and AI. The conversation opens with live Bitcoin price alerts showing sharp volatility, setting the tone for a broader discussion on the market drawdown, and new rules proposed in the EU and UK that tighten oversight on cash and stable coins.</p><p>The two explore why trust in central authorities is being tested and why gold and silver have suddenly diverged from crypto. They then shift to Coinbase’s attempt to reinvent the ICO era and the rising difficulty of launching tokens in a market that now holds more than 10 million tokens. From there Nish and Frackers move into the AI sector, debating whether AI valuations are sustainable after hundreds of billions have been spent on infrastructure and whether big tech’s decision to use debt for AI expansion is a sign of strength or impending doom.</p><p>In the markets segment, our hosts review last week’s predictions of the Bitcoin price, the altcoin season index and the fear &amp; greed index, compare their accuracy, and set new targets for next week — knowing they will probably be wrong.</p><p>The episode ends with the trading competition update, where Michiel’s diversified tech picks and Nish’s single-asset Solana position show very different risk profiles.</p><p>You can <a href="https://docs.google.com/spreadsheets/d/1eZ3tqwos9-w7p8Kz5xwkG9--6lDnpFWVNln0NxNIrso/edit?usp=sharing">follow the portfolio’s of Nish and Frackers here</a>.</p><p>They close with a discussion about long-term Bitcoin allocation, mainstream adoption and how early the industry still is. Oddly enough, both Nish and Frackers think Bitcoin will rebound next week. But they’re not so sure about the Big Tech stocks because of the ever growing AI-bubble.</p><h4>Remember: the NFA Podcast is <em>not</em> Financial Advice.</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9bf29928dcd5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Disaster week for Sam Altman, Mark Zuckerberg — and crypto]]></title>
            <link>https://medium.com/@frackers/disaster-week-for-sam-altman-mark-zuckerberg-and-crypto-c78501e930da?source=rss-bdc25f0e2591------2</link>
            <guid isPermaLink="false">https://medium.com/p/c78501e930da</guid>
            <category><![CDATA[altcoin-season]]></category>
            <category><![CDATA[sam-altman]]></category>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[crypto-trading]]></category>
            <category><![CDATA[tech-stocks]]></category>
            <dc:creator><![CDATA[Michiel Frackers]]></dc:creator>
            <pubDate>Tue, 11 Nov 2025 14:37:09 GMT</pubDate>
            <atom:updated>2025-11-11T14:37:09.100Z</atom:updated>
            <content:encoded><![CDATA[<h3>Disaster week for Sam Altman, Mark Zuckerberg — and crypto</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mxtX-X7AdxKcUwk1QqoVWg.png" /><figcaption>No, this chart is not inverted: these are the biggest crypto currencies over the last month.</figcaption></figure><p>It’s been a truly bizarre week in the world of Big Tech and Web3. While one Big Tech giant after another has published stunning quarterly figures in recent weeks, <a href="https://techcrunch.com/2025/11/08/is-wall-street-losing-faith-in-ai/">Wall Street wasn’t impressed</a>. Even though there was finally <a href="https://knowledge.wharton.upenn.edu/special-report/2025-ai-adoption-report/">a positive scientific study</a>, published by Wharton, not the shabbiest of business schools, which showed that the adoption of AI in the corporate world is largely successful and the AI boom appears to be continuing. Is Wall Street losing faith in AI? Could <a href="https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai">McKinsey’s report on the state of AI in 2025</a> have influenced Wall Street’s opinion about the business potential of AI-companies?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4FFjXHSsD1z_hl4cKYtM6A.png" /><figcaption>There has been little logic in the Big Tech stock prices over the past month</figcaption></figure><p>Tech companies that had otherwise performed well this year were among the hardest hit. Palantir’s share price fell 11% last week, Oracle dropped 9%, and Nvidia lost 7%. These declines follow Meta and Microsoft announcing their plans to continue investing heavily in AI, causing both companies’ shares to fall by around 4%.</p><p>Last week, the not-yet-publicly traded flagship AI company, OpenAI, seemed to finally be freed from its financing problems after reaching an agreement with majority shareholder Microsoft, allowing OpenAI to operate as a commercial company, likely <a href="https://frackers.com/en/openai-wil-historische-beursgang-van-1-biljoen/">leading to a historic IPO at a trillion-dollar valuation </a>. CEO Sam Altman was the man of the hour, but then he had a completely disastrous week, so now some cynics are <a href="https://polymarket.com/event/sam-altman-in-jail-by?tid=1762675647914">betting on when Altman will end up behind bars. </a>More on that later.</p><h3><strong>$360 billion in AI in a year</strong></h3><p>First, some numbers that the Washington Post neatly summed up. <a href="https://www.washingtonpost.com/technology/2025/10/30/google-meta-ai-data-center-spending/">Three hundred and sixty billion dollars </a>— that’s the amount the four richest companies developing AI (Microsoft, Google, Amazon, and Meta) spent last year building AI data centers and all the fancy gadgets in them, like the Nvidia chips the AI world needs.</p><p>All the announced or currently under construction AI data centers, operating at full capacity, would consume roughly the same amount of electricity as 44 million households in the United States. Big Tech’s profit margins are so high that this seemed easily feasible — with an emphasis on “seemed.”</p><p>Eight of the world’s top 10 most valuable companies are “AI-centric”: Nvidia, Apple, Microsoft, Google, Amazon, Broadcom, Meta, and Tesla. The combined value of these eight giants, $23 trillion, is greater than the combined value of the next ninety-six most valuable US companies, including JPMorgan, Walmart, Visa, and ExxonMobil.</p><p>The world’s most valuable company, Nvidia, last week became the first company in history to reach a market capitalization of $5 trillion. Nvidia’s value, today at $4.84 trillion, is still higher than the <a href="https://s2.washingtonpost.com/45b3c0a/690e2ca0aa7de408cb8499aa/undefined/29/53/690e2ca0aa7de408cb8499aa">entire stock markets in all but five countries worldwide </a>and approaches the combined value of all publicly traded companies in the United Kingdom and Germany. It’s realistic to assume that a Nvidia share price collapse would have ramifications for the global economy.</p><h3><strong>The AI value chain</strong></h3><p>The core of the problem is that many investors don’t seem to understand the AI industry’s value chain. <a href="https://www.instagram.com/reel/DQ0anq0kqC6/?igsh=MW9kdDcxMzBoZHMwMw==">George Hotz explains it very well </a>and offers this taxonomy:</p><p>Tier 5 Cursor, Windsurf (applications)</p><p>Tier 4 OpenAI, Anthropic, DeepSeek (large language models, LLMs)</p><p>tier 3 Nvidia, AMD, Broadcom (chips)</p><p>tier 2 TSMC, ASML, Samsung, Intel (fabs)</p><p>tier 1 land, energy, data centers, water (basic infrastructure)</p><p>Coreweave’s’s highly successful IPO <a href="https://www.google.com/finance/quote/CRWV:NASDAQ?window=YTD">(up 160%) </a>can be placed in tier 1 of this AI infrastructure framework, while the rest of the companies are well-known. The core of the problem Hotz outlines is that companies in tiers 4 and 5 have exorbitant costs, unlike those in the traditional web infrastructure value chain.</p><p>Sending an email with Gmail costs a fraction of performing a search with ChatGPT. Hotz therefore argues that the $200 he pays monthly for a ChatGPT pro subscription doesn’t even begin to cover OpenAI’s costs. OpenAI and Anthropic hope that their revenue growth will continue to rise so strongly that eventually the costs will be lower than the revenue. This hypothesis is increasingly being questioned, which is why the stock prices of Big Tech companies (Meta, Microsoft) that persist in investing hundreds of billions in AI are also falling.</p><p>The doubts about the structural profitability of all tier 4 AI companies like OpenAI are also the reason why OpenAI CEO Sam Altman is increasingly talking about future profits from a yet-to-be-released AI hardware device led by Apple legend Jony Ive, which I <a href="https://frackers.com/en/jony-ive-sam-altman-maken-iphone-killer-met-ai/">wrote about previously </a>. Altman, of course, dreams of the iPhone’s profit margins. Who wouldn’t in his position?</p><h3><strong>Horror week for Sam Altman</strong></h3><p>Altman’s week became a curious series of incidents. It began when, during a podcast with Microsoft CEO Satya Nadella and his own OpenAI investor, Brad Gerstner, Altman was asked the simple question how OpenAI would manage to order hundreds of billions of dollars’ worth of infrastructure in the coming years, while its reported revenue is “only” around $13 billion.</p><p>Altman <a href="https://techcrunch.com/2025/11/02/sam-altman-says-enough-to-questions-about-openais-revenue/">responded like a toddler throwing all his toys out of the pram.</a> Instead of explaining how OpenAI will finance its initial losses, which Gerstner clearly offered as an easy assist for Altman to dunk, Altman sneered: “<a href="https://youtu.be/Gnl833wXRz0?si=4F7GmL9gOdcU_KPk&amp;t=714"><em>Brad, if you want to sell your shares, I’ll find you a buyer.” </em></a>He could also have calmly replied that investors are lining up to pour money into OpenAI, as has been evident in recent years, and that the new corporate structure allows OpenAI to go public and raise tens of billions or more if profitability fails to materialize in the coming years. It was a weird moment.</p><h4><strong>Ten hours of testimony against Altman</strong></h4><p>Altman may have been so irritable because he knew he would be heaped with legal trouble later that week. In Elon Musk’s lawsuit against Altman, <a href="https://www.courtlistener.com/docket/69013420/musk-v-altman/">testimony was released </a>from a ten-hour interrogation of Ilya Sutskever, the highly respected co-founder of OpenAI, which painted <a href="https://decrypt.co/347349/inside-deposition-showed-openai-nearly-destroyed-itself">a disturbing picture </a>of Sam Altman’s reign. He was fired as CEO over a weekend in November 2023 for too often failing to remember the truth. The <a href="https://frackers.com/en/vijf-conclusies-na-de-chaos-bij-openai/">five conclusions I drew about the chaos at OpenAI </a>are still relevant and, pardon my immodesty, correct.</p><h4>Government guarantee for AI debts?</h4><p>The publicity surrounding Sutskever’s statement had barely died down when, on Wednesday, Sarah Friar, OpenAI’s CFO, said the following at a Wall Street Journal event:</p><p><em>“The backstop, the guarantee, that allows the financing to happen, that can really drop the cost of the financing but also increase the loan-to-value, so the amount of debt that you can take on top of an equity portion.”</em></p><p>Translated roughly from these accounting terms: <em>“If the US government would just guarantee all the debts that AI companies like OpenAI incur when financing their infrastructure, then we can really get this party started.”</em></p><p><a href="https://www.linkedin.com/feed/update/urn:li:activity:7392049356012507136/">In a statement on LinkedIn, </a>Friar subsequently said that by “backstop” (guarantee) she certainly didn’t mean “backstop” (guarantee). Nobody understood what she was saying so all hell broke loose. CNN’s headline read: “<a href="https://www.linkedin.com/feed/update/urn:li:activity:7392049356012507136/">Why OpenAI went into full crisis mode on Thursday.”</a></p><p>David Sacks, President Trump’s top advisor with the delightful title of “White House AI &amp; Crypto Czar,” <a href="https://x.com/DavidSacks/status/1986476840207122440">stated on X</a> that the US government will under no circumstances bail out the AI sector in the same way it once bailed out the banking sector after the mortgage crisis. This was yet another reason for the decline in share prices of Big Tech companies that continue to invest hundreds of billions in AI infrastructure.</p><h4><strong>Altman rediscovers capital letters</strong></h4><p>By now, it was clear that it was all hands on deck at OpenAI. Altman, who always boasted of never using capital letters “<a href="https://x.com/sama/status/1735123080564167048">because it reminded him of school</a>” apparently felt he had some homework to do and <a href="https://x.com/sama/status/1986514377470845007">published a 15-page memo</a>, with caps, in an attempt to calm the situation. Altman insisted that OpenAI doesn’t want the US government to provide guarantees for its data center projects.</p><p>Unfortunately… OpenAI did indeed <a href="https://decrypt.co/347866/openai-sought-government-loan-guarantees-sam-altman-denial">request government support for AI infrastructure in an eleven-page letter to the White House </a>. You don’t have to be a Marxist to see the hypocrisy in extolling the virtues of the free market as long as you can extract tens of billions from investors, only to ask for a government bailout if it turns out your entire industry is a pipe dream.</p><h4><strong>Seven charges against OpenAI</strong></h4><p>This week, it also emerged that OpenAI is facing a series of lawsuits in California. Families of victims in <a href="https://abcnews.go.com/Technology/wireStory/openai-faces-7-lawsuits-claiming-chatgpt-drove-people-127279676">seven lawsuits </a>accuse OpenAI of ChatGPT inducing suicide and delusions.</p><p>On the prediction market site Polymarket, <a href="https://polymarket.com/event/sam-altman-in-jail-by?tid=1762675647914">a contract </a>was opened trading on whether Altman would be imprisoned before December 31, 2025. Although the “yes” scenario has a very low probability, it’s striking that these sites have opened up Altman’s legal risks as a market.</p><p>Altman is dogged by lawsuits, even from <a href="https://www.bbc.com/news/articles/cz6lq6x2gd9o">his own sister </a>in a sexual abuse case, creating the image of a man who, as the Scottish proverb goes, can get into a fight walking into an empty house. Being served a subpoena by the San Francisco Public Defender’s Office <a href="https://www.instagram.com/reel/DQxTBGTESDJ/?igsh=bWJ0b3Q3ZXd6ZzZ4">while on stage in a theatre</a> with Golden State Warriors coach Steve Kerr, is painful for everyone. And it’s bad for business.</p><h3><strong>Meta has a lucrative scam problem</strong></h3><p>That other pale-skinned CEO who constantly gets himself into trouble, Mark Zuckerberg, also found himself in hot water this week when internal documents revealed just how big Meta’s role in the world of online fraud has become. A recent study estimated that ten percent of Meta’s advertising revenue, <a href="https://www.platformer.news/meta-scams-prevalence/">over $16 billion, comes from fraudulent ads</a>.</p><p>Advertisers with a slightly lower risk profile than outright scammers are simply charged extra fees, allowing Meta to profit even from questionable campaigns. A delightful business model for the amoral. The result: an ecosystem in which crime is profitable for Meta, as long as margins remain high enough.</p><p>Mark Zuckerberg had more on his mind this week: a huge neighborhood dispute has erupted over <a href="https://arstechnica.com/tech-policy/2025/11/mark-zuckerbergs-illegal-school-drove-his-neighbors-crazy/">the school he</a>’s built on his Palo Alto estate. In itself, the fact that Zuckerberg has established his own school is more positive news than the increasingly popular notion that he <a href="https://www.bbc.com/news/articles/cly17834524o">largely bought the island of Kauai </a>as <em>“doom prep</em>” for the impending end of the world. Yes, because of AI.</p><h3><strong>Crypto world experiences disastrous weeks</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mxtX-X7AdxKcUwk1QqoVWg.png" /><figcaption>The largest cryptocurrencies lost an average of 22% of their value last month.</figcaption></figure><p><a href="https://a16zcrypto.substack.com/p/state-of-crypto-2025-the-latest-data">Crypto may have become mainstream </a>in large parts of the world (Europe is lagging behind), but prices have been doom and gloom in recent weeks. “Uptober” was more of a “Floptober” and now it’s full on “sad November.” The chart above isn’t upside down; these are the crypto prices compared to a month ago.</p><p>In <a href="https://youtu.be/5vGNz9NitQU">episode 29 of the NFA Podcast, </a><a href="https://www.linkedin.com/in/dr-nisheta-sachdev/">Dr. Nisheta Sachdev and I </a>discuss one of the most turbulent weeks in crypto and AI. Markets are red — Solana down 20 percent, Bitcoin 10 percent — while institutional investors completely control the market.</p><p>The conversation takes a political turn when Nish points to on-chain data suggesting Eric Trump bought Solana just before his father publicly endorsed it. We then discuss two DeFi disasters: the $93 million loss at Stream Finance and the $128 million bridge hack at Balancer. Two painful reminders of how vulnerable DeFi still is. Nish explains how centralized exchanges are also under pressure, citing MEXC as an example, which admitted to mishandling the <a href="https://x.com/cecilia_hsueh/status/1984291339257647176">$3 million freeze from its client “the White Whale.”</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Goj1qo2i-5bCwHFsM18hZA.jpeg" /><figcaption>A trading competition in tech stocks and crypto seemed like a fun idea — back in June.</figcaption></figure><h3><strong>Next hype: quantum</strong></h3><p>After all the commotion about too much hype in AI and crypto, the <a href="https://www.bbc.com/news/articles/c04gvx7egw5o">first doubts are now being raised about quantum computers</a>. It would be nice if we can ride out the AI and crypto waves before labeling the next wave of innovation as madness.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c78501e930da" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>