Join a group of 56,000 followers on @SeekingAlpha, ranked #6 amongst all subscription services on the platform.
Signup for a 20% discount to Out Fox The Street through the end of June to get premium daily updates, live portfolios & much more. $AMD $NBIS $SOFI $HIMS
https://t.co/QMhHjQ6hRd
College students affordability issue is all about not knowing how to cook or settling for a simple sandwich. Lived off PBJ sandwiches and a banana in college. Kids these days want Chipotle and to eat out multiple times a day whereas I only ate out Friday nights.
Last night, my wife made homemade burritos for our family of 8 and it cost less than $20 total for everyone.
Im beginning to think the affordability crisis is really an “I don’t know how to cook” crisis
$INVZ Innoviz Tech. recently entered the Physical AI race with a $3.5M order from a defense tech. company. The Lidar company announced 6 other engagements plus the robotaxi business is finally gaining traction. The company is finally stacking revenues, but liquidity is a big issue here with Physical AI revenues only targeted at ~$7M, or 10% of $70M revenue base.
♦️Six New Engagements in Defense and Security – Kela, Drive Group, Regulus, Givon Defense, Cogniteam, and AeroNous
♦️Liquidity as of June 30, '26: ~$48.5 million + $30M equity offering in July
@realEstateTrent yes, and most people won’t look at stock positions after tax trapping them into holding the stock forever due to the perceived net worth hit.
$VZ $SPCX Verizon investors were warned of the upcoming threat from Starlink Mobile. The whole US wireless industry of AT&T, Verizon, and T-Mobile got another warning from the SpaceX President & COO Gwynne Shotwell on SpaceX Q2'26 earnings call:
"Between AT&T, Verizon, and T-Mobile, there is roughly $600B/yr between them. I anticipate us acquiring a few of their customers, because I think our service will be better — we will eliminate dead zones by leveraging the satellites in orbit. It will be better during any natural disaster. We will start providing service at the end of next year."
@WalksWallstreet you’re confusing the project with where billionaires live. Ken Griffin still lives in Miami and will likely shift more work to Florida.
Can someone help bridge me? cc: @elonmusk
How does the 10GW comment makes sense other than as an aspirational target?
SpaceX disclosed ~1GW of installed nameplate compute across Colossus 1 and 2 as of March 2026; the Q2 call puts that at 1.4GW. Epoch shows ~1.3GW currently installed, which is directionally consistent.
Aterio identifies 1.07GW of construction scheduled to activate in 2026: Minihard (200MW), Google lease (150MW), Macroharder (500MW), and Colossus 2 Phase 2 (220MW). That bridges current capacity to ~2.4–2.5GW, which lines up with Elon’s >2GW year-end target.
The Texas projects do not help in 2027 as both are only announced, with listed activation dates in June and September 2029. So 10GW by end-2027 requires another ~7.5GW of terrestrial capacity beyond the disclosed pipeline?!
Orbital cannot fill that gap absent a major change in timing. SpaceX’s own initial public offering prospectus says orbital compute deployment begins “as early as 2028.” Even if it were pulled forward, SpaceX’s stated orbital assumptions imply an extreme scale requirement: 100kW of compute per metric ton and 100 metric tons per Starship launch. Filling a 7–7.5GW gap would require 70–75k metric tons of satellite hardware and 700–750 Starship launches. The prospectus itself says its 100GW annual orbital ambition requires thousands of launches per year.
On terrestrial cost, Epoch puts Colossus 2 at $35.8bn for 946MW and Colossus 1 at $12.9bn for 340MW: both imply ~$38bn/GW. At a rounded $40bn/GW, the missing 7–7.5GW requires ~$280–300bn of gross capital that isn't contemplated right now in sellside figures?
That could ultimately be financed by SpaceX, suppliers, lenders, partners or third-party developers. But the sites, power and funding are not visible in the disclosed 2027 pipeline. The remaining possibilities are large third-party powered-shell leases, partnerships with other hyperscalers, or acquisitions of existing front-of-meter or behind-the-meter capacity, including crypto sites. Even then, most would still require data-center fit-out and execution within roughly 17 months?
Somebody help me out.
ZEE, our aviation specific AI foundation model, has made a frontier breakthrough.
It has demonstrated the capability to generate highly accurate predictions of real-time aircraft trajectories on the airport surface minutes ahead of the present. With runway-related incidents accounting for 30% to 40% of global aviation accidents, this ability to provide a highly accurate ‘window into the future’ gives the humans in the loop the most critical asset in aviation safety: time to react.
Live testing is underway at Hawthorne. The early results measured against real-world tracking data have been strong, and larger-scale testing is now underway.
More details: https://t.co/BedpkEEwpK
Read the ZEE Technical Blog: https://t.co/wXgXguTSeP
$UPST Upstart trades sharply higher as Q2 earnings climb, helped by core personal loan growth. Loan originations were only $4.2B, atill not topping the 2022 peak. https://t.co/ColieXqqA4
Sorry but $AMD numbers this Q are pretty mediocre.
They beat for sure but nothing insane.
$NVDA continues to beat by larger percentages and is a much lower valuation
Far better mega cap semi play IMO