$NEBIUS(NBIS)$ $IREN Ltd(IREN)$ $APPLIED DIGITAL CORP(APLD)$ $SanDisk Corp.(SNDK)$ $Micron Technology(MU)$ There's an interesting pattern forming that keeps reminding me of the summer 1998 dotcom melt-up. Back then, dotcom stocks were booming and took a huge correction in summer 1998, with a major fund, Long Term Capital Management, blowing up. Almost 30 years later, we have an AI boom, a big pullback this summer, and now a major fund, Situational Awareness, has blown up. The similarities between the two periods are striking
$NEBIUS(NBIS)$ $IREN Ltd(IREN)$ $APPLIED DIGITAL CORP(APLD)$ $Cipher Mining Inc.(CIFR)$ $TeraWulf Inc.(WULF)$ AMZN posted some pretty strong numbers that look good for infrastructure companies broadly. AWS grew 36.7% year-over-year in Q2, the fastest in 18 quarters. Their AI and Chips businesses each hit run rates above $25 billion. Revenue came in at $200.6B versus estimates of $196.4B, up 20% year-over-year. EPS was $5.75, which included a $53.4B non-operating gain mostly from the Anthropic investment. AWS sales reached $42.2B against an estimated $40.
$NEBIUS(NBIS)$ $IREN Ltd(IREN)$ $APPLIED DIGITAL CORP(APLD)$ $SanDisk Corp.(SNDK)$ $Micron Technology(MU)$ So maybe there's still a bit of hope left. Looking at the current market compared to the Dotcom Boom, there were plenty of insane pullbacks back then too. I remember the January 3 2000 pullback in particular. I dumped a lot of good positions that week, which turned out to be a terrible move — everything flew right back up and went much, much higher over the next 9 weeks.
$CoreWeave, Inc.(CRWV)$ Bears are acting like compute demand is just going to disappear. Honestly, wait until 2028. At these levels, IREN, NBIS, and APLD all look like a steal, even if they drift lower. Nothing fundamental has changed, so if we do see more downside, I'm just adding more.
$NEBIUS(NBIS)$ CapEx is the issue. But the market knows Nebius isn't spending money on Frontier models and runs Open Source options on Token Factory, right? Frontier models get commoditized and the Mag7 just wasted a bunch of money. But Nebius didn't. AI demand isn't going away, and Nebius has been skating to where the puck is headed — Open Source.
$NEBIUS(NBIS)$ Markets aren't rational and tend to overshoot on both the high side and the low side. The AI space is going through a correction right now, and a lot of great companies are down 30-50%. NBIS is a long-term hold for me, and I expect volatility over the next few years as it marches toward new highs.
$NEBIUS(NBIS)$ Just something to keep in mind — if Meta and the other big tech names are all chasing this, there's probably serious money behind it. I'm staying long and staying strong. No stock moves in a straight line.
$Applied Optoelectronics(AAOI)$ , $NEBIUS(NBIS)$ , $Micron Technology(MU)$ , DRAM.X have been taking heavy hits lately, coiling through the $81.54–$88.61 demand region. Feels more like a brutal shakeout than the whole AI theme actually popping. While some are throwing in the towel, it looks like smart money is aggressively sweeping liquidity off key support. A reversal pivot could be setting up here.
AI infrastructure is still one of the strongest themes in the market right now. $APPLIED DIGITAL CORP(APLD)$ just reported earnings with a massive revenue beat, over 152%. The AI race is clearly expanding beyond chips. The next phase looks like it's about the companies building the data centers, power, and computing capacity behind this whole buildout. $APPLIED DIGITAL CORP(APLD)$ seems to be benefiting from accelerating AI data center demand, but execution and margins will be the real test. $NEBIUS(NBIS)$ keeps drawing attention with its AI cloud infrastructure exposure as GPU demand keeps growing. $CoreWe