Bullpen.News From Edwin Dorsey
Get Daily Recap Feedback

Sunday, October 18, 2026

Back to today

From X

@convequity 4 clicks

Convequity on X

We think hardware manufacturing and chip costs, more than launch, will decide how much $SPCX earns from orbital compute. Our SPCX model estimates how much orbital compute can be made and launched, what it costs and what it earns as it ages, then adds that revenue to our forecasts for SpaceX's launch, Starlink and Cursor businesses in our SPCX DCF. Each year, deployment equals the gigawatts of solar and AI computing hardware the supply chain can make for orbital data centers, capped by what Starship can carry. In our base case, launch is the tighter limit only in the first four years of the forecast, and manufacturing is the limit after that. Our base case puts chips and racks at about 80% of the cost to build each gigawatt and launch at about 11%. Solar arrays, satellite structure and laser communications make up the rest. Because chips make up most of the cost and launch only a small part, whether SpaceX uses merchant GPUs, custom chips or makes its own chips at Terrafab will matter more to returns than how cheap launch gets. Revenue per gigawatt depends on the business model, and we assume it is higher for selling AI services through an API than for renting out compute. Revenue falls as the hardware ages, and we assume each satellite is replaced after a 7-year life. The model's price-cut settings let us test price competition. We think Chinese companies could use their own launch and computing infrastructure to offer cheaper AI services internationally, pressuring the prices U.S. AI companies charge and what they can pay for capacity. The U.S. and China could also tie access to their AI services to defense and trade relationships. Countries with ties to both, including India and some Middle Eastern states, could keep access to U.S. and Chinese providers and negotiate lower prices.
@BrokenMoats 2 clicks

Broken Moats on X

Have to at least acknowledge rate reversal and move into low quality and retail/software today...... But oil/diesel/soft commodities/transportation all staying bid, find this more positioning over extended om rates then durable change (especially with the data and the still unresolved issues causing the inflation) AI sold off I think is what is driving both moves in rates (more out needs a home) and retail/rate sensitive/software were the funding shorts (see sectors and components today) ......but geez at these prices an easier trade I think is to fade software/retail then guess next stop on AI and funding.....the are correlated (for now) anyways. Go through the sectors and components and sort by % gain today almost stacked ranked from worst business (biggest intraday gain) to quality at the bottom.....tells you whats going on. Some might say low quality always leads on a reversal.....this is not a start this is after a stealth sharp move higher for $ARHS $ETSY $EBAY $KSS $SHAK $WMT $CBRL $APPN $DOCU $IT $FRSH $BSP etc.
@StableBread 2 clicks

Fajasy on X

I tried to work out who’s behind the AI data labeling companies that approached Nubeva Technologies $NBVAF (TSXV: $NBVA). CEO Randy Chou didn’t name the data labeling companies (firms that turn experts’ knowledge into training data for AI labs) or the customers behind them. “If a data labeling company’s coming to us, somebody behind them is asking,” Chou said, adding that the customer typically isn’t limited to cybersecurity companies. OpenAI is one potential candidate, since my previous write-up concluded that the frontier model Nubeva got access to in September was most likely OpenAI’s, through its Daybreak program for vetted defenders. OpenAI launched GPT-6 Astra on September 3, 2026, the day Nubeva announced its access. It’s the first model OpenAI rated Critical for cybersecurity capability. But access to OpenAI’s model wouldn’t prove OpenAI is also behind the data labeling companies, since any lab building cyber models could want labeled malware. Among data labeling companies, three of the largest are publicly recruiting or selling binary analysis expertise, the skill of figuring out what malware programs do: → Scale AI: Its product manager posting calls the cybersecurity portfolio “a product line that does not exist yet” and says the hire will recruit “malware analysts.” → Turing: Sells CyberBench, security tasks for training and testing frontier models, with malware analysis and reverse engineering among its ten categories. TechCrunch called Turing “a key coding provider for OpenAI” in March 2025. → Mercor: Recruits security experts, preferring reverse engineering or binary analysis backgrounds, and counts OpenAI, Anthropic, and Meta $META as customers. OpenAI cut ties with Scale in June 2025, after Meta bought 49% of Scale and hired its founder, Alexandr Wang. So if OpenAI is behind the approach, Turing and Mercor are the likeliest data labeling companies, since both sell to OpenAI and already sell or buy binary analysis expertise. Either way, these data labeling companies are worth $2.2-29B, so even Turing, the smallest, could pay the top of my $12-19M sale range for Nubeva for less than 1% of its last valuation.
@convequity 1 click

Convequity on X

AI swarms could make network security more important. Anthropic's October 12 research ran an AI model across 15 cheap Android TV boxes at 0.39 tokens per second. The small models still couldn't reliably complete the full replication cycle. I think we're still early. Inference on this hardware is painfully slow. But the idea is interesting because human-level intelligence isn't necessary for something to reproduce and spread. Think of insects: limited individual intelligence, yet populations can spread widely. An AI worm could carry its model with it, adapt to unfamiliar devices and use compromised hardware to run the next copy. A model too large for one device could be split across several. Better processors and more memory in future cameras, TV boxes and other connected devices would make more local inference possible, wherever attackers can gain access. That creates a security problem beyond installing software on individual devices. You're not going to put an endpoint security agent on every camera or TV box. You need to identify those devices through the network, restrict what they can communicate with and isolate them when compromised. A perimeter firewall alone cannot inspect traffic that never passes through it. I think $FTNT could benefit here. Its agentless device discovery and network segmentation fit this problem, especially across distributed sites. More capable AI on unmanaged devices could increase demand for those controls. The opportunity also extends to networking vendors that integrate security into switches and wireless access points, where they can control how devices connect and communicate.

Investor Letters

All letters

2026.42: Home At Last

Stratechery’s weekly bundle covers Apple HomeView and SpaceX’s cellular ambitions.

AAPL $5T
Stratechery by Ben Thompson – On the business, strategy, and impact of technology. 6 clicks

TSOH Weekly Roundup (10/16/2026)

Bumble agreed to a $1.3 billion take-private deal, while Netflix cut about 1,000 jobs and Meta discussed its Muse superintelligence models.

NFLX $511B
TSOH Investment Research 11 clicks

Is The Market Turning The Page?

CoreWeave’s debt and equity price action is framed as a wake-up call for investors amid a broader market discussion.

CRWV $34B
WHAT ARE WE DOING? “Contrarians At The Gate” 52 clicks
LLY · long

LILLY (ELI) & CO

Eli Lilly trades at 33 times this year’s earnings and 26 times next year’s, versus roughly 12 times for big pharma peers.

LLY $1.2T
Value Investors Club 2 clicks

AppLovin Deep Dive

AppLovin rebuilt its ad model with AXON 2.0, lifting LTM revenue to about $7 billion and margins to 77% as Unity competition threatens…

APP $88B
Speedwell Research 27 clicks

Constellation: Stars Aligning?

Constellation Brands trades near $127 at roughly 11x next-twelve-month EV/EBIT, with recent business results offsetting concerns over…

STZ $22B
TSOH Investment Research 27 clicks

TSMC Q3 FY12/26 Earnings Review

TSMC reported excellent Q3 results, guided Q4 revenue growth higher and saw its shares slip despite strong AI-driven earnings momentum.

TSM $2.3T
Cestrian Capital Research 9 clicks

Publications we follow

Every publication on this site, hand-picked.

130 publications

Newsletters 108

YouTube 4

Blogs 17

Seeking Alpha 52 of 240

Nominate a publication

X accounts we follow

Every account on this site, hand-picked.

89 accounts
Nominate an account

Investor letters we follow

189 firms

Sectors

Ordered by how often our editor reaches for them.