Trump's First Week: Inauguration Recap, Executive Actions, TikTok, Stargate + Sacks is Back!

Watch on YouTube ↗ | January 25, 2025 at 02:59 | 2:05:33 | All-In Podcast

Speakers

David Friedberg — CEO, The Production Board

Thomas Laffont — Managing Director, Coatue Management

Jason Calacanis — Angel Investor / Founder, LAUNCH

Chamath Palihapitiya — CEO, Social Capital

David Sacks — General Partner, Craft Ventures

Summary

The All-In hosts recap Trump's inauguration weekend and his record first-day executive orders with guest Thomas Laffont of Coatue, then turn to markets: how to value TikTok's US business if Washington takes half of it, and whether the $500bn Stargate announcement can ever earn a return. The most investable stretch is a long segment on electricity, where charts show the US far behind China on generation capacity, cost and nuclear additions, leading to a case for nuclear power and a specific mention of Constellation Energy's behind-the-meter contracts with hyperscalers and federal agencies. Chamath reads Netflix's share-price chart against rising SSRI use and falling labor participation, while Laffont teases Broadcom and Hock Tan's M&A record against Intel's governance failure. David Sacks joins at the end straight from the Oval Office to describe three new executive orders on crypto, AI and PCAST.

Thomas Laffont values TikTok's US business at roughly $100bn using Meta's daily actives and time spent, and does not think a trillion is unrealistic long term.
Chamath argues Washington should take equity or royalties when it grants permits, incentives and loans, using the Dior deal and Tesla's repaid DOE loan as templates.
Stargate's $500bn is debated: Laffont says the money exists if the ROI does, while Chamath says the DeepSeek release shows AI costs collapsing and the number is mostly marketing.
US electricity capacity, price per kilowatt hour and nuclear additions are presented as the real constraint in the AI race with China.
Constellation Energy is named as the public-market nuclear play, with Amazon, Microsoft and federal agencies contracting power directly behind the meter.
Friedberg says Gen 4 reactor technology already works and only regulation blocks a US buildout, calling for emergency deregulation.
Chamath frames Netflix's rise as a leading indicator of societal disengagement, while Jason says he is a happy shareholder.
David Sacks details executive orders creating a crypto working group, rescinding the prior AI order and re-establishing PCAST.

Ideas

David Friedberg CEO, The Production Board 83:41

Only regulation blocks a US nuclear buildout.

Friedberg argues that electricity, not chips, is where the US loses the AI race to China: the US pays roughly 1.5 to 3x China's price per kilowatt hour, has about half of China's production capacity, and China's cost to add a new gigawatt is only one tenth to one half of the US cost, so if AI and automation are the drivers of growth the US is structurally disadvantaged. He says the technology problem is already solved - Gen 4 meltdown-proof designs such as the pebble bed reactor are already producing electricity in China while the US deploys zero - and that the only blocker is regulation, with plenty of capital, intellectual capital and talent eager to build. He wants emergency-style deregulation, calling the power shortfall a bigger national security threat than the border.

Thomas Laffont Managing Director, Coatue Management 88:17

Nuclear is the binding AI race constraint.

Laffont frames the grid as the binding constraint on AI: a GPU is useless without a data center and a data center is useless without power, and his charts show US nuclear capacity essentially flat for more than 25 years while China ramped hard, with France as proof that a developed economy can run about 70 percent of its electricity on nuclear. He expects the trend toward nuclear to keep going and to be a key part of the AI arms race over the next 10 to 20 years, and adds that denying China advanced GPUs will not matter if China ends up with roughly ten times the power capacity, because sheer industrial AI capacity would outweigh chip quality.

Thomas Laffont Managing Director, Coatue Management 88:37

Constellation benefits from behind-the-meter nuclear deals.

Laffont points to Constellation Energy as the public-market way to own the nuclear leg of the AI buildout. His differentiated evidence is contracting behavior rather than policy talk: Amazon and Microsoft are signing power agreements directly with nuclear operators in behind-the-meter structures to secure dedicated electricity for data centers, and late last year federal agencies also began contracting directly with CEG for their own facilities. Even though that federal deal was small in absolute size, he and the market read it as an acknowledgement, under the prior administration as well, that nuclear is the best path forward for firm large-scale power.

Jason Calacanis Angel Investor / Founder, LAUNCH 96:03

Holding Netflix; subscriber strategy is working.

Calacanis says he is a Netflix shareholder and is delighted with the position: subscriber growth looks like it is back on track, and even the much-criticized live fight event appears to have ended up driving the stock higher, so in his read the current strategy is working even if he cannot fully explain what management is doing right.

Chamath Palihapitiya CEO, Social Capital 96:45

Attention-sink businesses win as people disengage.

Palihapitiya's opposite take is that Netflix is a very well run and highly performant business whose success is a leading indicator of American disengagement. He overlays the stock's decade-long rise with the rising share of Americans on SSRIs and a falling labor force participation rate, and argues the company wins by being a sink for people's time, amplified by algorithmic personalization that linear television never had: people get hyper-tuned content, binge for a weekend and do not work, exercise or see friends. His investment implication is that the more people detach and opt out, the better these distraction businesses do, which is good for the stock even if he thinks it is bad for society.

Thomas Laffont Managing Director, Coatue Management 104:58

Broadcom compounds on Hock Tan's M&A mastery.

Laffont raises Broadcom as the name he wants to dig into next: a business that began as a roughly $3.5bn double spin-out of HP and was compounded through a long series of roll-ups into a trillion-dollar-plus company. His differentiated point is that this was not one winning product but Hock Tan's M&A mastery and portfolio construction, buying the right assets ahead of the curve and integrating them, which is exactly what Intel failed to do over the same stretch.

Thomas Laffont Managing Director, Coatue Management 107:24

Intel's governance failure caused lasting underperformance.

Asked why Broadcom compounded and Intel did not, Laffont's verdict is that Intel really missed the curve and that this was a complete abdication of corporate governance at both the board and the CEO level. The comparison on the table was Broadcom's market cap up about 8x and Nvidia's up about 20x over five years against Intel down roughly 70 percent over the same period, a divergence he attributes to the failure to buy and invest ahead of where semiconductors were going.

Chamath Palihapitiya CEO, Social Capital 109:50

Winning AI requires many more nuclear plants.

Reacting to the US versus China generation-capacity charts and to the new AI executive order, Palihapitiya says the whole country now accepts it has to win the AI race and that, on those numbers, the only way to win is to build many more nuclear power plants. He expects the new AI and crypto czar to hammer this through and frames the political resistance to it as a fight the opponents lose.

Up Next

‘Enormously Dysfunctional Catastrophe’: Economy Could Be 'Destabilized' | Richard Wolff

The David Lin Report · Jan 24

$TRUMP Coin Insanity! w/ Qiao Wang

Steady Lads Podcast · Jan 24

Trump's $500 Billion 'Stargate' Is 'Most Important U.S. Investment' Today | Arthur Herman

The David Lin Report · Jan 24

Get Ready For '100k Job Cuts in January' | Danielle DiMartino Booth

The David Lin Report · Jan 23

What Trump’s First 100 Days Will Do To Bitcoin, Cryptos | Galaxy's Steve Kurz

The David Lin Report · Jan 23

This All-In Podcast video, published January 25, 2025, features David Friedberg, Thomas Laffont, Jason Calacanis, Chamath Palihapitiya discussing URA, CEG, NFLX, AVGO, INTC. 8 trade ideas extracted by AI with direction and confidence scoring.

Speakers: David Friedberg, Thomas Laffont, Jason Calacanis, Chamath Palihapitiya · Tickers: URA, CEG, NFLX, AVGO, INTC