AI reindustrialization (software/hardware/robotics) = top priority for government & corporations. Investing in front of change as the process unfolds…
RoadTo10Figures
Joined September 2019
- Tweets2.3K
- Following1.4K
- Followers3.1K
- Likes3.3K
Cash With Dara retweeted
UKRAINE'S ZELENSKIY SAYS UPDATED INTELLIGENCE SHOWS RUSSIA IS PREPARING A 'MASSIVE ATTACK'
ZELNSKIY SAYS MUCH DEPENDS ON REACTION OF WORLD LEADERS WHO 'HAVE THE DEFENSIVE CAPABILITIES WE NEED AND CAN PROVIDE THEM TO US'
Identity will continue to shift towards wellness and health.
Health is wealth is a quote I hear a lot.
It’s a trend I see a lot too with those who are successful, don’t need to work, and look to other things for actualization.
The promise of AI is currently framed as relieving humans of the tedious tasks that encompass work, allowing them to focus on the more rewarding and creative problems of novel discovery and idea generation.
This has been mostly true, but as AI gets better I think it has a high likelihood of becoming a comforting lie that we tell ourselves. The more capable AI becomes, the more humanity will have to resist being relegated to the middle management layer of its own progress. And the larger the risk that all human work becomes machine-like without any real sense of fulfillment or accomplishment.
As AI gets better at the stuff that goes beyond execution and into the realm of judgement, creativity and purpose, the boundary between AI work and Human Work is going to destabilize. Without a conscious effort to push back against this, it’s easy to see a future where humans are relegated to supervising systems they no longer fully understand.
“Drudgery” is defined by repitive, tedious tasks. But perhaps a better definition for it would be work with the absence of agency, and under that definition AI can certainly create more of it. Plenty of coders have already experienced a loss of meaning in their work as AI “solves” coding.
“What is joy without despair” is accepted to be a fair question. But what is self-actualization without the overcoming of friction or frustration. I do worry a bit for the future generations that may live an entire life without experiencing the qualia of success after repeated full-effort attempts met with failure.
If AI can execute our ideas, it’s likely that eventually it will be able to create better ones. Mastery and ownership of a creation being things of the past is hard to imagine.
The most valid pushback to this concern is that people still play chess despite engines existing, make furniture from hand regardless of IKEA etc.
I wonder if we’ll see an investment theme over the next decade that has to do with the “meaning economy”, or whatever people buy when their work is no longer their identity.
Maybe in the more near term, there’s a trade in the value of accountability. If the vast majority of work isn’t done by a human, who do you yell at / fire when the work is bad? Anyone selling the right to blame a human or to outsource accountability (maybe E&O insurers?) could benefit.
The diffusion of ai means the world is becoming more capable.
level: base of knowledge has increased
slope: knowledge accrues faster now
curve: the speed of knowledge accruing is accelerating
Now the bottleneck is implementing the knowledge in the real world.
1/
This has occured in finance before. The US has pioneered a lot in finance field. When you implement that knowledge base into other countries for reward (foreign PE), you do it through a JV w someone local. It works the same with AI and humans, and liaison is the FDE.
4/
Slower is better if it means you are more intentional and make less silly mistakes
Slower is worse if you are path dependent and levered to speed
Delicate balance between do it right and do it fast (considering we are in a race)
Limiter is physical: humans & resources
Sam Altman admits he was wrong on AI's timeline and says society and the economy will adapt more slowly
"I thought when we got to GPT-4, which was back in 2023, that very quickly after that there was going to be much more disruption, software businesses up for grabs right away, than it turned out to be."
"I think I was wrong about a few things, but one in terms of the speed: the economy just has so much inertia."
"People keep doing the same things, buying from the same company, wanting to use their tools the same way. I think this is actually a positive in many ways, and it's going to make this big transition go smoother and slower. I'm grateful for it."
"But it means we've all been too ambitious on timelines. Even with this incredible technology, society and the economy will adapt more slowly."
Cash With Dara retweeted
The bottleneck in AI reindustrialization is humans building fluency w the tools
I think he is right about the sub of labor & proliferation of apps, j lacks perspective to predict max adoption for the masses bc hard for him to empathize w avg person bc he is already fluent
The copper and cat converter strippers will go crazy over these
Nvidia and a homebuilder want to bolt $150,000 of GPUs onto new American houses.
The box is the size of an AC unit. It's liquid-cooled, runs without fans, and bolts straight onto the exterior wall. Inside sit 16 Blackwell server GPUs, 4 server CPUs, and 3TB of RAM, which puts over $150,000 of hardware on the side of the house at roughly $10,000 per GPU.
The homeowner never buys any of it. Hosting is the whole arrangement, traded for subsidized electricity and internet, a smart panel, and a backup battery, usually against a flat fee of around $150 a month.
Span, the startup behind the program, keeps full ownership of the GPUs and sells the compute to AI cloud providers. The pilot covers 100 new homes in the Southwest, and the stated target is 80,000 nodes pushing a gigawatt of compute by 2027.
Here's why the deal exists at all. Seven in ten Americans oppose a data center near them, and New York and Texas have paused permits for the large ones. A box on a private house skips that entire fight.
The suburb quietly becomes AI infrastructure. The compute value flows to the companies, the homeowner gets a discount on their bills, and the rules for compute on a private house haven't been written yet.
Would you take that trade?
So things are going very poorly… North Korea risk factor is back into play.
North Korea has been preparing for conflict for a few years now and are a credible threat.
This is a sensitive time for the US so to give this concession now is telling.
It’s too early to call the bottom on software
Just because AI giga longs were cleared out doesn’t mean that they were wrong about the medium term and that software fundamentally improved
Dispersion is evident but the proliferation of AI is still so early to make convicted calls
During the internet boom comprehensive revisions incorporating software capitalization, improved hedonic price indexes (quality/feature adjustment), and fuller source data revised and raised average productivity, revealing gains some had already inferred.
Booms in new types of technology have a habit of improving productivity. But measuring that improvement isn’t easy and can take a bit of time.
In the 1990s, software, personal computers and the internet were expected to raise labor productivity growth, but the data didn’t show that. At least not at first.
In 1996, Alan Greenspan looked at the numbers for labor productivity growth. They were low. He asserted that the current data were not yet reflecting the full benefits of the information and communications technology boom. As it turned out, over several years, those low productivity numbers were revised higher, bringing the official data closer in line with the accelerated productivity Greenspan had predicted.
ALFRED is the companion to the St. Louis Fed's FRED database that helps you see the data that were available at points in the past and how they were revised over time. Here’s what the data showed:
➡️ In September 1996, labor productivity growth looked like it averaged just 0.9% from 1989 to 1995.
➡️ In February 2000, after a major revision to the National Income and Product Accounts in 1999, it was raised to 1.4%.
➡️ As of June 2026, it stands at 1.5%.
Visit the FRED Blog for the full story and the data bit.ly/4vX976H
Friends help friends 🤝
The Trump Administration delivers for America's trusted partners. Economic security is national security. And the U.S.-Japan alliance is built on both.
Friday's coordinated foreign exchange actions countered disorderly yen movements.
Treasury remains attentive and in close communication with our counterparts at MOF and BOJ. We will not hesitate to participate in further joint intervention.
The FIMA Repo Facility is an important backstop. We would encourage it to be upsized in the coming months.
We strongly support Japan's decisive market and monetary steps to correct the substantial undervaluation of the yen.
The Takaichi government is moving into an exciting new phase of Abenomics, as nearly 15 years of powerful stimulus have created durable, robust underlying economic dynamics.
*
The framework for the U.S.-Japan alliance begins with the strong relationship between @POTUS @realDonaldTrump and Prime Minister @takaichi_sanae.
We've been in close contact with our Japanese allies, and they are making serious efforts to address the substantial undervaluation of the yen. Through our conversations, we believe they will continue to put the right policies in place to lead the yen’s return to a more normal equilibrium price.
Currency intervention is a signal that governments want to change a direction. But only strong, effective policies will result in sustained yen strength.
*
Japan to cut off the supply of advanced packaging equipment to China starting August 1st.
On August 1, Japan's third round of new export controls on semiconductor equipment officially came into effect, including core equipment for the entire advanced packaging chain in a list of strict controls. The regulations adopt a case-by-case approval model, and with a long approval cycle and an extremely high rejection rate, they effectively cut off the supply to the domestic market.
This marks the third major upgrade to Japan's semiconductor export controls on China, focusing on addressing the shortcomings previously addressed primarily in front-end wafer manufacturing equipment and bringing core equipment across the entire advanced packaging chain under strict control.
The first round (July 2023): Japan will include 23 categories of advanced semiconductor manufacturing equipment under export controls, including cleaning, deposition, heat treatment, photolithography, etching, and inspection equipment, mainly targeting equipment capable of producing logic chips at 14nm and below.
Second round (2025): Further tighten and expand the control list, strengthen control over photoresists, materials and catch-all supplements.
The third round (announced in May 2026, effective August 1): On May 29, 2026, the Ministry of Economy, Trade and Industry (METI) of Japan revised the relevant list of the Foreign Exchange and Foreign Trade Act, and for the first time included a complete set of back-end packaging and testing equipment for advanced packaging (Chiplet/3D stacking/CoWoS/TSV) in the export control list.
The newly added key controlled items (approximately 20 categories, with core components related to back-end packaging and testing) include: high-end die bonders, ultra-thin wafer thinning/grinding machines, laser cutting/hidden dicing machines such as DISCO, TSV (Through Silicon Via) equipment, bump fabrication/micro-bump plating equipment, high-precision chip sorting machines, hybrid bonding machines, flexible bonding equipment, and related high-end testing equipment.
Previous sanctions imposed by the US, Japan, and the Netherlands mostly focused on front-end wafer manufacturing equipment such as lithography and etching machines. This round of restrictions, however, has pinpointed China's weakness: advanced packaging equipment.
Today, advanced chip manufacturing processes are approaching physical limits. To produce high-performance computing chips and HBM high-speed memory chips at low cost, chiplets and 3D stacking packaging are the most feasible breakthrough paths. Ultra-thin wafer thinning machines, laser slicing machines, high-end die bonders, and TSV silicon via equipment are the hardware foundation of this route, and the market has long been monopolized by companies such as Japan's DISCO.
According to the new regulations, all new orders for this type of equipment to China will require separate export licenses, with an approval cycle of 3-6 months. The rejection rate for applications for advanced process equipment is close to 80%. Existing orders signed before August 1st will only have a one-month transition period for fulfillment. After the transition period ends, new machine procurement and original manufacturer maintenance services will be completely discontinued.
