@ParadisLabs

If you wanna feel more at ease: Many of my high conviction AI names are down significantly today. $NBIS: -17.1% $GLW: -13.6% $AEHR: -12.1% $ALAB: -10.8% $MU: -10.6% $SNDK: -10.6% $LRCX: -9.7% $INTC: -9.0% $MRVL: -8.7% Just for a small subset of names that I hold, and plan to keep for the forseeable future. Nice to see them get kinda offset by $META, and software type names like $RDDT and $NOW though which I hold a decent chunk of. Also, it's pretty interesting to see the inverse correlation between semis and software recently. Defo something to note if anyone else takes short-term swing trades too.

Paradis Research | $META the neocloud Meta may build a cloud business to sell excess AI compute capacity. TLDR: - $META selling excess compute because they literally have no roadmap to use old gen capacity - This is not bearish for neoclouds like $NBIS - $META would instead dump old Hopper/Blackwell it overbought while renting new Vera Rubin capacity from $NBIS for 2027. Just to frame Meta's own scale since it matters for what "excess" actually means: - Meta has guided 2026 capex to ~$135B, up ~75% YoY. - They've said they'll buy "millions" of $NVDA Blackwell + Rubin GPUs this year alongside Grace CPUs. - While simultaneously scaling MTIA, their in-house training/inference accelerator designed w/ $AVGO. Two implications: 1. At that spend, even a low % of "overbuilt" capacity is a multi $billion pool of resellable compute. 2. A cloud business is years from positively impacting Meta's rev mix which is still mainly from ads. But naturally, the market has sold off the neoclouds today out of fear: - $NBIS -14% - $CRWV -13% However, the threat to the incumbent neoclouds has been massively overstated ever since the news came out. I've been cringing reading bearish posts. Per $NBIS March release, the two signed a 5 yr deal worth up to ~$27B: - $12B of dedicated Vera Rubin capacity delivering early 2027 - Up to $15B of flexible capacity that Nebius intends to sell to third parties, with Meta buying whatever is left over as a backstop. So Meta is simultaneously renting frontier compute from Nebius and floating the resale of its own. Pretty funny lol. Meta wants guaranteed compute allocation speed which Nebius with their Nvidia partnership can deliver. While the capacity it would resell is older generation kit (Hopper, early Blackwell) it over ordered. That segmentation dissolves most of the paradox and probably tells you the resale pool is marginal + ageing supply. Naturally, the reflex reaction is to think that Meta will crush GPU pricing + breaks the compute shortage thesis: If Meta tip from net renters (Microsoft rents from $CRWV and Nebius exactly because they're short on capacity) to net sellers, the marginal price of rented compute softens, and the neocloud and neoclouds get impacted. But the bulk of the neocloud revenue is contracted and pre-paid over multi year deals. Nebius's ~$50B backlog and CoreWeave's own book anchored by Microsoft insulate them. And bitcoin miners like $IREN, $CIFR, $CORZ, $WULF, $GLXY etc sit further out on the risk curve, with more spot exposure. Across the basket, Meta's resellable excess is just a marginal addition to supply rather than a flood like some will have you believe. Then if Meta prove internally that selling excess compute is more profitable than ads... They'll almost certainly increase capex next year to Google/Amazon levels to go build a cloud biz in full.

Super interesting seeing everyone jumping in on the humanoid trade with $CCXI recently! Just important to remember that it's still extremely early: - Est. 2026 AI capex = ~$1 trillion - Est. 2026 humanoid VC funding = ~$70 billion Personally, I'm investing early as a core believer in robotics / humanoids. But I do kinda feel like you'd want to wait and see which OEM like Agility or Figure breaks out to become the de facto leader and go-to OEM as prodcution volume ramps up. For B2B customers like $AMZN, $MELI, and Toyota etc. Similar to $NVDA and GPUs. At that point, the upstream supply chain starts to open up where you'll see names like Harmonic Drive, Leaderdrive etc also have steeper volume/financial inflections especially if supply constrained (highly dependent on China). Personally, I've been waiting years for humanoids to enter the commercialization phase and wanna participate in that early. Just still very early to build conviction with the likes of Siemens / $TSLA still running proof of concept tests in a small number of factories. With a tail of other OEMs like "Humanoid" in the UK a couple gens behind the likes of Agility. And you've still got issues where it's taking some time to actually develop humanoid hands for more tricky tasks on top of moving boxes in warehouses etc. So then you've got a further commercial TAM expansion into fields like medical/research on top of logistics currently.

Gemini and $GOOGL might be cooked chat. https://t.co/yiinPlt24v

Yeah, I've got positions in $CCXI for exposure to Agility. I do believe that they're one of the best humanoid producers globally ex-China. Especially their current Digit v4 commercial humanoid which has one of the best spec sheets worldwide. Some important points on Agility's BOM: -> Actuators are ~50% of the BOM for a humanoid -> Schaeffler are crucial here for Agility. Schaeffler: - invested in Agility - buy Agility's robots - but are also Agility's strain wave/planetary actuator merchant supplier And Schaeffler are probably the best actuator platform worldwide (maybe more than Harmonic) since theirs are smaller and lighter than rivals. And, going back to the BOM, Agility get the actuators at cheaper rates than rival OEMs like Humanoid (a UK humanoid maker...obviously lol). Preferential supply treatment etc. So w/ economies of scale: As more customers like Amazon etc buy from Agility -> the cheaper the actuators become -> the cheaper the humanoids become = Agility in theory become the de facto humanoid leader. If Schaeffler ofc maintain supply agreements and preferential treatment (which they can as an Agility investor). And looking at Agility's order book w/ 30+ customers in the pipeline, it does seem like their BOM has been rapidly declining recently. Unlike memory w/ $SNDK and $MU, you do want humanoid costs to come down for scalability and commercial reasons. And has been deployed in: - GXO Logistics - Schaeffler - Toyota - $MELI - $AMZN With basically every deployment citing "multi site" language for the future beyond current single site deployments. Which Agility have plenty of capacity for at ~10k units per yr if you wanna believe the CEO vs. hundreds of units currently being produced. Amazon for example explicitly need costs to come down e.g. that's why they're trialling robotics in Poland rn. For payback analysis mainly. Which if that goes well, you can expect to see wider deployments in the US in the coming years. Same with $MELI in Latam too. That being said, I do think SPACs are inherently risky - I think we all know the risks. But given I'm a humanoid enjoyer, I do wanna participate in some way since that's where the next AI paradigm shift will be beyond "software".