Jo Bhekid
67 posts
SpaceX, as predicted (not just by me) continues its slide.
There are a lot of share unlocks incoming starting in August, continuing selling pressure.
HOWEVER, at $1.6T, we are moving into risky territory for bears. According to our analysis (coming out this weekend in Pioneerlands.org), SpaceX is likely to add significant EBIT THIS YEAR.
Significant means a run rate of up to $20B. That would put the stock at a 50 fwd P/E - which would be criminially cheap for a stock that can grow earnings 100%+ for a very long time.
Of course not as criminally cheap as $MU or $NVDA, but with much larger moats.
The full story is of course complicated - BUT at 1.5T and 30B EBIT run rate, SpaceX would have been transformed from extremely expensive to very cheap within a few months.
In the Age of AGI, things move VERY fast.
AJ Investment Research@alojoh
SpaceX has now managed to defend an over $1,000 billion valuation for 27 sessions! Stock currently still at $1,617 billion. 👏
English
That seems insane .
Have to look into this.
$NVDA is my third largest position and I remain very bullish regardless
Gavin Baker@GavinSBaker
Interesting report from @theinformation that Nvidia will be able to make 1000 Vera Rubin racks per day, which is $630b per quarter. Actually a little hard for me to believe and haven’t checked the math, but wow if true. And Vera CPU racks and Groq LPU racks would be incremental to this. As would their new business model where they take a revenue share of neocloud revenue in return for guaranteeing offtake. I do think the latter point has not been well explored by analysts and is likely to be super important going forward.
English
70k miles a year is not too optimistic - it’s roughly 200 miles a day, which is equivalent to current ride share cars.
$1 / mile would definitely do the trick in the US and many other markets (but kit Poland , Eastern europe etc - I agree). Average $/ mile for Uber is something like $2.5
So the math works out for the US
kolowski@kolowski
@JOBhakdi Cybercab will cost $20,000 per year to operate. If you want to earn 50k, it needs to make 70k miles charged at $1 per mile. This is very optimistic. In Poland, a human operated taxi costs less than $1 per mile 🥴.
English
AfD doesn't need coalitions in Sachen Anhalt, Meck Pom, Sachsen and Thüringen if they get a little lucky - they are close to absolute majorities.
On the federal level, the only option is the CDU - unless AfD surges to 43%+, which is very possible imo.
If CDU signals that they are willing to form a coalition, that could stop the bleeding. They could do so once Merz steps down, likely in Q4.
If they refuse, they will descend even more into chaos and lose more members. Together with the rapidly increases crisis levels in Germany, that could push the AfD towards 43%+ next year.
The unraveling can happen faster than people think.
Germany is in a ridiculously bad situation that is completely self-inflicted to the extent that it looks like sabotage:
Merkel opened the borders, then - against German law - allowed millions of foreigners to stay and drain the treasury and justice system, then they (literally) blew up Germany's power plants, allowed the US to (literally) blow up their most important pipeline, committed heinous crimes throughout COVID, and now (literally) prepare a war against Russia, while also jailing thousands of political dissidents.
JUST IMAGE a US government that
Gives the order to open the border (ok that actually happened, BUT:)
THEN make millions of illegal migrants legal,
THEN BLOW UP all American nuclear power plants, THEN allow the UK and France to blow up the Keystone Pipeline (which isn't even close to the importance of Nordstream) and
ALSO jail everyone who complains about it for "hate speech"
Do you think people would get angry?
Nietzsche@Nietzsche183684
@JOBhakdi Who is AfD making a coalition with u think?
English
The US trying to ban (or create opaque liegst threats) to CH insane open source models would be an enormous strategic mistake.
Let’s see what happens
Bindu Reddy@bindureddy
Wow - woke up to some really bad news The nightmare scenario is coming true - the US may ban open source models Europe and China will dominate the global AI market and kill the US - If this happens
English
That would qualify as a massive planning failure, and that never happened before with $TSLA. That is my simple argument. Once you go above 10k or so Cybercabs, you move into the 200M+ misallocation area.
Guessingpanda@guessingpanda
You never quantify why they can’t go longer. Of course they can. Production of 100k cabs is 2bn dollars. They could go a year without deployment. No law of physics or economics prevents it. All it does is push ROI to two years instead of one for the launch phase. Always look closely at the other side of your argument.
English
What is fascinating to me is that Fable, Sol 5.6 etc are good at excel work, coding and other things - but are catastropic failures in writing English.
They score Fs and Ds on the most basic things (when you write a summary or essay, start with context, not in the middle of some autist stuff), and then keep failing at every element of decent English (sentence structure, vocabulary etc.), not to speak of higher-level stuff (argument structure etc.).
Why?
Isn't basic writing important across all kinds of fields - any why is it so hard? Should be very trainable.
English
This market is in a pretty horrible mood.
Iran is looking more and more like a problem that won't go away - the US seems incapable of bailing out of it, and China etc. won't be able to contain Iran forever. They still have a huge incentive to escalate, and only a clean cut US retreat would prevent that from happening.
Then - the S&P is still near ATH, even though Semis and Ai stocks are getting slaughtered.
What should be an easy buy is therefore still at risk, because a market correction can still happen any time.
Very tricky.
English
Yep, markets are super tough right now - important to put things in context when looking at individual stocks
$TSLA is holding up relatively well , probably because it has underperformed by a lot
$SPCX continues to fly too high imo, valuation is hard to defend in the current environment. Long term it's obviously a no brainer, but I don't see how it can hold up over the next 6 months with more sell catalysts coming.
$NVDA and $MU, as well as other semis, are now ridiculously undervalued but short term momentum is against them.
I am still holding back with $SPCX, probably until Nov.
$TSLA can surprise with Robotaxi and probably WILL, over the next 6 months, but we don't know when. Risk is that we see a market crash in Q4, which would drag it down.
$MU, $NVDA etc. - it's hard to see how these won't be MUCH higher from here very soon.
Ivan@Aganin7
@JOBhakdi Thanks Jo! Markets have been so tough. Hope you are looking after yourself and sleeping well.
English
Inkling - the next US based open source frontier model .
Looking better than I expected .
Commoditization is happening quickly - huge pressure building on OpenAI and Anthropic by year end
Ejaaz@cryptopunk7213
what an absolute huge win for open source AI in america. 1st ever model from OpenAI's ex CTO and its a 1T open model that reasons across text, code, images, video - but the best part isn't even how good it is... its how you can fine-tune it to create something way better: > you can adjust the thinking effort > you can influence the way it approaches a coding problem > you can perfect token spending so your task costs pennies this solves a big problem that companies are facing now: they're spending $100Ks - $10Ms when they could just be fine-tuning a "less intelligent" model to do the same task at a fraction of the cost (and often better) thats why bridgewater's using Thinking Machine's tinker product i expect to see a lot of folks playing around with this model, finally an alternative to chinese open source!
English
The common man better get organized quickly because the overlords won’t care much about him.
The 2030ies will be all about civilizational restructuring on a massive scale.
Those who use their own brain and understand reality -
AND create networks with others who do so -
AND who pick people they actually like and are willing to create shared value with - will be the survivors.
It’s going to be a time of great chaos and opportunity, and it starts … now
Dave Hawkins@Permieguy
@JOBhakdi What do you think will be some of the most impactful effects of all this space compute in the common man?
English
Dumping heat is a totally solved problem. They do it all day long , no pun intended , with Starlink sats.
Radiation: shielding and computational compensation are solved problems too. There is some degradation, but effectively that just lowers compute output over time.
The advantages of no permitting, lower install costs (including launch costs ) and lower operating costs easily offset this.
Asymmetric Edge Research@AsymEdge
@JOBhakdi Powering a Rubin rack is only half the challenge. In space you still have to dump all that heat without air, protect the chips from radiation, and keep everything running with nobody there to repair it.
English
We are clearly in a momentum moment and I wouldn’t bet on a specific number for $MU bottom.
But long term, this looks insanely undervalued already
Elon Era@Elon_Era_ai
@JOBhakdi Downside target?
English
