New Ask Claude or ChatGPT about any stock, using Bullpen. Bullpen in Claude or ChatGPT. 7 days free
Bullpen.News From Edwin Dorsey
Get Daily Recap Feedback

Sunday, November 15, 2026

Back to the front page

From X

@aleabitoreddit

Serenity on X

New commentary today from SoftBank on Elon Musks post 4 days ago: “SoftBank is defaulting on their turbine orders” Analyst Question: “Any comment on that?” SoftBank CFO Yoshimitsu Goto: “this is simply not correct at all. There is no such default” This also mirrors what $GE CEO Scott Strazik stated about seeing no procurement cancellations or counterparty financing difficulties. Pretty damaging allegations about the company financing $SPCX competitor (OpenAI + Stargate). I wonder if there’s any more nuance these claims or if it’s false.
@aleabitoreddit

Serenity on X

If you look at CW laser ASPs for AI data centers, the pricing differences are pretty substantial: 70mW: ~$3–4 100mW: ~$4–6 (Source Photonics channel checks). With some premiums apparently going up to $7-$9 range with the recent shortage for pluggables. 120–150mW NPO: ~$15–20 350mW UHP CW: ~$40 (Nomura) 400mW UHP CW for NPO/CPO: ~$50–75 (Mizuho) With former communication laser chips going from 20-50 cents. The pricing and gross margins of these players go up with each architectural shift (60%+ for CPO lasers). And go up even more depending how far down the stack you go (selling ELSFP was ~2.5x revenue content) In terms of how I see the landscape now: - $COHR is probably the best risk adjusted laser player now, for people who don't like uncertainty. Their new $20B LTA is very substantial. - $LITE pretty negative now given the valuation overhang from $MRVL acquisition. Operationally very opsitive. - $AAOI has extreme demand visibility (likely $6-$9B in new LTAs from just 1 quarter), but credible execution uncertainty. They've also been focusing on pluggables + bottlenecked by DSPs, rather focusing on UHP CPO laser merchant markets. But I think if markets see a new $5B-$7B LTA that they're in late stage discussion for... and see the $4.4B MC. Pretty sure that's a catalyst for recovery. The demand doesn't disappear but timeline shifts back a few months. AOI also signaled DSP as the main bottleneck, last quarter it was TIAs, so other different component players that are interesting. - $SIVE has extreme capacity expansion (300m laser capacity targeted in total). The economic implications that they won't disclose are likely enormous relative to their MC, even if they sell $3 70mW lasers to China pluggable players. But there's a qualification uncertainty to fill the gap, that markets have punished the company for. Especially when management communication amplifies uncertainty, despite the ongoing shortage. Right now entire AI trade is going up and down together though.
@kevin__mak 2 clicks

Kevin Mak on X

The post below dovetails into my view on "powered shells" or essentially the utility companies of the AI trade ($CORZ, $RIOT, $WULF, etc.) These companies largely (or entirely) own no GPUs. They lease a building with electricity/cooling/infrastructure to the hyperscalers/neoclouds/AI labs, who bring in their GPU racks/fiber and run their datacenters. They're the bottom ~$1 of the "per GPU hour" value chain. They're essentially real estate/utility company plays where they collect pre-specified rent for a very long time. As far as I can see, the only real downside that these companies face is that their tenants go bankrupt, AND nobody else takes over the lease/continues running the datacenter. In other words, datacenters "go dark" due to overbuilding. In my opinion, this is exceptionally unlikely. The cost of running these facilities is exceptionally small relative to the neocloud or "frontier model" pricing. If margins ever compress, it should come out of the "top of the stack" way before it comes out of anywhere else. In a world where there's an overbuild of datacenters, and datacenters are literally dark, it also means new ones aren't being built. That means GPUs and memory and all that infra are no longer being sold, because it's way easier to turn on a dark datacenter than it is to build a new one. If new datacenters aren't being built, most of the semi/hardware industry is also dead. These powered shells are occasionally trading below their contracted value, never mind adding the terminal value of their buildings and the option value of their expansion capacity. I think the reason why they're trading down so hard lately is because they're "correlated to the AI trade". But the mistake the market is making is that their operations are much less risky than the rest of the "AI trade" stack. The heroic assumption for current pricing isn't "industry grows at 50% yoy", or "margins stay >40%". It's just "datacenters don't go dark", which is a very low bar. I own $CORZ and $RIOT because I think you're getting pretty close to a free call option on AI growth. As long as these datacenters don't go dark, I think you make your money back plus a small/reasonable return (depending on your assumptions). If the AI sector grows and leases more power, renews leases, etc., then these companies will benefit. Disclosure: Creek Drive and its affiliated funds holds long positions in RIOT and CORZ. Views expressed are current opinions for information only, subject to change and are not investment advice or a recommendation to buy or sell any security. Investing involves the risk of loss and past performance is not indicative of future performance. See the disclosures link in my bio for more important information.

Investor Letters

All letters

Q3 Earnings Updates

Q3 updates cover Crocs, Entravision and Fiserv beats, coal and oil weakness, and Xerox’s sharp decline, testing the underlying cases for…

CROX $7.5BEVC $874M
Hidden Rock Capital's Newsletter 3 clicks

The Spy Who Shagged The Market

Nvidia’s leadership scandal puts its AI-market role under scrutiny, with comparisons to AIG’s pre-GFC liquidity influence.

NVDA $3.9TAIG $43B
WHAT ARE WE DOING? “Contrarians At The Gate” 4 clicks

Is Nebius Fucked?

Nebius’ Q3 beat relied on $325 million of one-off partnership fees as it raised 2027 CapEx to $40–45 billion and may need $6–11 billion of…

NBIS $32B
Daniel Romero 6 clicks

Qualcomm Custom ASICs?

Qualcomm’s Alphawave SerDes, CPU cores and HBC could support custom AI ASICs as hyperscalers build heterogeneous fleets beyond GPUs.

QCOM $185B
Chipstrat 4 clicks

Publications we follow

Every publication on this site, hand-picked.

130 publications

Newsletters 110

YouTube 4

Blogs 15

Seeking Alpha 88 of 295

Nominate a publication

X accounts we follow

Every account on this site, hand-picked.

91 accounts
Nominate an account

Investor letters we follow

376 firms

Sectors

Ordered by how often our editor reaches for them.