Thursday, August 27

Thursday, August 27, 2026

Everything we published on this day.

61 stories — the front page that day →

@KEDM_COM

KEDM.com on X

Versant $VSNT, a recent Comcast spin now seems to have found its way up. Weak price action though since the spin on classic spin dynamics, a ‘melting ice-cube’ narrative, and (we think) mainly index selling (it isn’t in the S&P500). A reminder that Versant owns CNBC and some other networks. With a $5.3bn market cap, roughly $2bn net debt, and $1.0-1.5bn free cash flow generation per year, the ‘bet’ here is that you will get your money back in 5-6 years. Imagine if FCF stabilizes. And Versant seems confident, launching a $1bn share buyback, almost 20% of the current market cap.

UK Primer: The Equity Revival

UK equities trade at decades-wide discounts to developed-market peers as improving sentiment, capital reform and foreign M&A support a…

@negligible_cap

Negligible Capital on X

*META COULD SPEND UP TO $10B/YEAR ON ANTHROPIC AI: NYT So $META is like 15%+ of Anthropic's ARR? "At one point this year, Meta internally projected that it could spend as much as $10 billion annually on Anthropic’s A.I. models, said two of the people, who declined to be identified discussing private information. That would have formed a major chunk of Anthropic’s yearly revenue, which the start-up estimated in July would pass $65 billion." - NYT
@negligible_cap

Negligible Capital on X

70% FY28 revenue growth is incredible, but if you were an $NVDA bear before that call, it’s very unlikely that view has changed (sure maybe funding shorts get covered) The bear argument is more intact than ever. Nvidia’s demand, while very robust, still comes in part from largely unprofitable companies/labs which are of course dependent on Nvidia for financing Jensen addressed these concerns directly, saying "frontier labs sales are skyrocketing, their margins are fantastic…they're generating profitable tokens…they're only limited by the amount of compute” The bear argument isn't going to be solved by fantastic numbers in any given quarter, it will take lots of time to be proven right or wrong
@RagingVentures 1 click

Raging Capital Ventures on X

I agree there are differences in HBM vs. NAND. I said we are “around” the peak of the cycle, with $SNDK much more at risk. Memory is an intensely cyclical business with a long history of booms and busts. Cyclical peaks are typically associated with double and triple+ ordering, as we are seeing today. Paybacks on new capacity are 1 year or less; capacity will come. $ASML is the primary limiter of capacity growth for DRAM/HBM, which is bullish for limiting supply growth in those markets. NAND is far less defensible. Memory is the biggest pain point in the ecosystem today and many entrepreneurs and engineers are intensively focused on solving this problem with innovation and other adjustments. De-speccing is already commonplace and memory pooling is coming. Finally, I am not saying it is time to short these stocks (although I am short some $SNDK). What I am saying is it is very poor capital allocation for $MU, $SNDK, etc. to be buying back stock at these multiples of invested capital.

The Verra Sneerer (NASDAQ:VRRM)

Verra Mobility renewed Hertz tolling management for five years. AutoKinex lets rental customers buy tolling, fuel and protection products…

GM 2026-08-27

General Motors: The Billion Mile Moat

General Motors trades at 8.2 times forward earnings as it targets $25 billion of software revenue by 2030 through OnStar, Super Cruise and…