1. Podcasts
  2. This Week in Startups
  3. Why a Do-Nothing Congress could be great for markets
This Week in Startups

5 Nov 2026 · 1h 25m · 29 chapters

Episode artwork

Why a Do-Nothing Congress could be great for markets | E2359

In short

Venture capital roundtable on the implications of a “blue wave”/split government for markets and tech—especially data centers, defense/robotics, self-driving, AI economics (margins, GPU/power constraints), and M&A timing. Guests also discuss how AI adoption may change public narratives and consumer sentiment.

Guests (backgrounds)

  • DA Wallach, co-founder of General Partner Time BioVentures; venture investor/LP.
  • Ilana Gold, from Two Roads Capital; runs an investment syndicate doing deal-by-deal investing in AI, infrastructure, space, robotics, defense, and energy; has invested $50M+ across 200 companies.
  • Tomáš (last name not provided), founder/general partner at Theory Ventures; early-stage (seed to Series A) investing, $5M–$50M investments.

Key claims

  • Divided government may reduce “crazy” policy swings, which markets like; uncertainty is a bigger driver than AI “crowding out.”
  • Data centers are politically salient but power constraints (e.g., “run out of power in 2026”) will shift innovation toward chips/nuclear and efficiency.
  • Anti-AI/anti-automation narratives may lose as consumers experience real benefits (agents, robotics, faster services).
  • AI margins are likely to compress due to price wars, open-source, and staffing reductions (e.g., Canva cutting ~22%).
  • GPU/power scarcity will ration “all-you-can-eat” inference and increase app-layer innovation.

Notable examples

  • Polymarket/polls vs election outcome; NBC exit poll: 56% cite AI/data centers as major vote factor.
  • Panthalos ocean nodes; Waymo/FSD safety discussion; Zipline delivery; Canva staff cut; Marvell/JDS Uniphase deal; Sequoia scout program mentioned.

Written by AI. May contain mistakes.

Chapters

0:45

Introducing Guests and Their Backgrounds

Discussion introducing DA Wallach and Ilana Gold, their experience and focus areas.

“We're an investment syndicate, so we do deal-by-deal basis.”
2:28

Impact of Split Government on Tech

Analyzing the effects of the recent election results on venture capital and technology.

“We have a lot to get to here, and we got to just talk about the most amazing 48 Hours DA.”
8:23

Challenges Facing Data Centers and Autonomous Vehicles

Exploring the narrative around data centers and the implications for self-driving technology.

“And so what startups are going to build to be like, hey, maybe we can't build out as much data centers, but maybe we're going to build chips that don't require as much power.”
14:03

The Tension of Self-Driving Cars

Discussing the implications of self-driving technology on jobs and society.

“of anti-self-driving go across the country.”
16:47

Intergenerational Challenges in the Workforce

Examining the economic divide between young workers and retirees regarding technology.

“Because if they do take this position that young people or socialists want them to take, well, now we're getting into Ayn Rand kind of territory.”
19:37

The Role of AI in Everyday Life

Exploring how AI can help individuals and small businesses improve efficiency.

“And that was, I think, one of their efforts.”
25:19

AI's Potential for Political Engagement

Discussing how AI could improve public understanding of political issues.

“I think there's the digital domain and then there's the physical real world, right?”
28:00

Exploring Federalism and Local Solutions

Discussing how federalism allows for innovative local solutions to big problems.

“You've seen obviously in Texas where you are, how much they've been able to move the needle on this in cities like Austin or Houston.”
30:00

AI's Impact on Daily Life

Exploring how AI and robots can improve everyday tasks and the potential societal benefits.

“Zipline, no delivery fee or the delivery is a buck or it's incorporated into the price is essentially what I think will happen.”
32:27

Driving Automation and Insurance Dynamics

Analyzing the relationship between self-driving cars, insurance, and safety improvements.

“I think maybe car insurance ultimately ends up looking like title insurance.”
Show all 29 chapters
34:37

Economic Implications of AI Margins

Discussing the sustainability of high margins in the AI industry and competition concerns.

“Your competitive advantage period wanes.”
41:19

Market Dynamics and AI Pricing

Examining the market pressures affecting AI pricing and sustainability.

“And my question is, well, then why don't the prices just go up?”
42:00

Pharma vs AI: Time to Market

Explore the differences in commercialization timelines between pharma and AI companies.

“If it takes me half a billion dollars or a billion dollars to make a drug, if I'm a pharma company, I have 17 years to commercialize that with a patent.”
43:10

The Reality of Hardware Companies

Discusses the challenges hardware companies face in maintaining profitability and market share.

“Like, as the market came down, I just bought a million dollars in Uber and Tesla, and I totally hit the bottom.”
44:18

Local Inference and AI Accessibility

Examines the trend of local AI models and their implications for companies and consumers.

“Like Coca-Cola's gross margins and net margins are not zero because they're competing with Pepsi.”
46:06

Evolving Business Models in Tech

Analyzes the need for hardware companies to maintain profitability while adapting to new tech trends.

“some of the editing that's made for your podcast, some of the preparation.”
50:18

Changes at Sequoia Capital

Discusses recent changes at Sequoia Capital, including the retirement of their scout program and restructuring efforts.

“So thanks to Fintrix for supporting the show and supporting my syndicate, syndicate.com.”
54:09

The Future of Venture Capital

Explores the implications of Sequoia's changes on the venture capital landscape and the importance of adapting to new investment strategies.

“funds have become incredibly complicated businesses.”
56:00

The Evolution of Venture Capital and Scouts Programs

Exploration of the changing dynamics in venture capital and the role of scouts.

“seed or you put but 100 or 250K, and we were putting 25 and 50K checks in originally, then it went to 100, 250.”
58:34

The Rise of Allocorns and Profitable Startups

Discussion on allocorns and the shift towards profitable startups avoiding traditional funding.

“profitable, DA, which is why I popped up the barbell strategy with the syndicate.”
1:00:38

Challenges Facing Venture Capital Firms Today

Analysis of the obstacles faced by venture capital firms as markets evolve.

“Well, I think this is, again, this is like the great tension of this moment.”
1:02:54

Access to Venture Capital for All Investors

Advocacy for broader access to venture capital opportunities for non-accredited investors.

“And they're the ones who are going to have a really good shot at getting this sort of alicorn class.”
1:10:03

Investing Accessibility for Everyone

Discussing the potential for broader investment access in startups.

“An Uber driver could participate in investing in Uber because they were driving Uber and they saw how popular and how much the customer love was.”
1:10:55

The Case for Earlier IPOs

Exploring the reluctance of companies to go public despite significant revenues.

“And if you go, who has a huge voice and you want to go raise publicly, suddenly now you're not just an accredited investor.”
1:12:20

The Impact of the Jobs Act on Public Companies

How regulatory changes have allowed more companies to stay private longer.

“I'm sorry, if you've got a$5 billion company that's doing$800 million in revenue, why can't you be public?”
1:14:32

The Rise of Tokenized Assets

Examining how tokenization is changing asset trading dynamics.

“Explain what it is, because I don't know if everybody understands what high-tech.”
1:16:40

The Future of Trading with Perpetual Options

Understanding how perpetual options can simplify trading and increase accessibility.

“If you really wanted to get a stock, a share of SK Hynek, it's actually really tough.”
1:19:07

Risk Taking in Today's Economy

Discussing the implications of high-risk trading behaviors among investors.

“then that would make the existing assets go down in value if too many people move over here?”
1:24:07

Future of Writing in Tech

Discussion about the future of writing and the role of humans in tech.

“As you were saying, I'm hiring a lot of people.”
A man walking through an airport, asking his phone a question

Ask about this episode

Get the gist, explore an idea or ask about something said.

Claude

On This Week in Startups, what are the key points of this episode?

Searching your listening

Transcript

Automatic transcript. May contain errors.

0:00All right, everybody, welcome back to This Week in Venture Capital. It's our Venture Capital Roundtable for November 4th, and we have a lot to talk about. Before we do anything, I have my Plod pin on. I'm going to start recording so I don't miss a minute. If you are involved in a lot of conversations like I am, you can record all those conversations, interviews, meetings, calls. Check it out at plod.ai slash twist. Use the code twist to get 10 % off, and they know that we sent you. All right, back for a second week in a row. My guy, DA Wallach, is here, co-founder of General Partner Time BioVentures.

0:36I'm an LP. How are you doing, DA? I'm doing great. Awesome. Great to be here. Great to have you here. For the first time, Ilana Gold from Two Roads Capital. Tell us a little bit about Two Roads. We're an investment syndicate, so we do deal-by-deal basis. We're in AI, Infra, space, robotics, defense, and energy. Some companies have done is like General Compute. They just raise their Series A, Panther Lassa, Allo Atomics, XOI, et cetera. But super excited to be here, Jason. Been a huge fan of you for a while. Oh, well, thank you. That's very kind of you to say. And you've done over 50 million across 200 companies.

1:13Do you use AngelList as a platform? Or is the syndicate over there if people wanted to check it out? Yeah, AngelList and Sidecar. But you can find me if you just look up Two Rose Capital, AngelList. It should come up. And that's my previous syndicate in this one. And yeah, it's been super exciting. Yeah, Sidecar is a great platform as well. And Tomas is back, the founder and general partner at Theory Ventures. Man, you and I have been at this for a long time. You're on your second fund or third fund?

1:40Tomasz Tunguz: Second fund. Yeah. And mostly early stage Series A seed still? Seed in Series A, yeah. So anywhere from about$5 million investment to about$50 million. This Week in Startups is brought to you by G2I. G2I has been recruiting software engineers for over a decade. Go to g2i.co slash twist to find out how G2I vets engineers and can help you hire them in days, not weeks. Vanta compliance and security shouldn't be a deal breaker for startups to win new business. Vanta makes it easy for companies to get a SOC2 report fast. Get$1 ,000 off for a limited time at vanta.com slash twist. Every. For all of your incorporation, banking, payroll, benefits, accounting, taxes, or other back office administration needs, visit every.io.

2:27All right. We have a lot to get to here, and we got to just talk about the most amazing 48 Hours DA. We had a lot of folks hand-wringing, oh, the prediction markets are wrong. Shout out to our partners, Polymarket. Oh, the polls are wrong. This is all made up. They're all fake news. But a blue wave happened. It wasn't a spritz. The Democratic Party is taking control of the House and the Senate. We now have a split government. And some of these issues are pretty significant for tech. Obviously, the data center trade being the number one trade. Overnight markets look like we were going to have the market correct 2%, 3%, 4 % today.

3:19And instead, we've had the opposite. And I think the theories there, DA, are, hey, the war is going to end. Oil will be having a battle of madness, and that kind of chaos will go away. Great state of Texas was a clean sweep, shockingly, for the blue. So don't blame me for that. That's not just because I came here. But DA, what's your take on what this means for the venture capital startup, entrepreneurial community, and just the wider country? Yeah. Well, look, for the wider country, democracy works. It's a great thing. This is what we want. We want this vacillation between the parties. We don't want a monarchy.

3:59We don't want one-party rule. We want there to be an active debate in the country that's happening on Main Street and that's happening in Washington. And I think that's exactly what we've got as it pertains to our worlds of venture capital and the capital markets. Look, I think it's got two sides of the coin, right? On the one hand, obviously, the data center story is a big political winner, and it's probably going to be a big political winner for both parties. So I think whether, you know, I think the narrative for the Democrats this time was we're going to stop these data centers. And, you know, in a couple of years, the narrative on the other side, maybe we're also going to stop these data centers.

4:41So, you know, like what politician wants to get up there and say, we've got this great technology we need to support that's coming for your job. And whether it's intellectually true that AI ultimately displaces jobs and ruins local communities and takes too much power and too much water, whether any of those things are true, what matters is that you have to jump through all kinds of hoops and contort yourself to make the counter argument in other words for mark andreessen to go through the whole history of technology and point out with exacting detail how every other wave we thought was going to put out everyone to work but then it didn't that's a more complicated story even if it's true so you got to read like five or six books.

5:30Yeah. And I mean, it's, you know, have two or three long duties. Yeah. Now the flip side, maybe why we're seeing the markets not be so negative, even though that could really take the thunder out of AI is that Trump has been a chaos agent, right? And, you know, whether he likes that or it's just something he can't help but do, he sows this endless level of uncertainty. And capital markets do not like uncertainty. So I would maintain, we don't need to get too into macro. I would maintain what we've seen happen in long-term interest rates over the past year has been much more about uncertainty and fiscal unsustainability than it has been about crowding out with AI borrowing or anything like that.

6:17And when the uncertainty around everything in macro goes up, markets just don't like that. So I think divided government, even if it means sclerosis, even if it means that nothing gets done in DC, what it does mean is that nothing crazy is likely to happen. And that's good for business. Nothing getting done is good for business. It's like a little bit of indigestion there. And they're debating stuff endlessly in committee. It means they can't also start a war. It means They can also start tariff chaos makes things a bit more predictable. Ilana, one of the things that the Trump administration was great at the last few years was embracing the tech community, American dynamism, you know, defense tech and supporting data centers, supporting the industry, perhaps even too much.

7:08And at the expense of talking about and optics do matter, the plight of the, you know, everyday American. What do you think this will, how this might impact what you're doing in terms of defense tech, in terms of American dynamism, robotics, the hardware? A lot of these things have government as customers, and you might have a government that is less excited about having these founders to the White House. Yeah, I think one of the biggest narratives are happening right now is people are saying, okay, like no more build out of data centers. And I think it's going to be obviously a state issue now of, hey, if you're building out data centers, are you going to be able to build them out nationally?

7:54And that's a really big question. But I think that still doesn't change the underlying fact that we don't have enough power right now. I think Elon just tweeted this past week saying like, we're going to run out of power in 2026. So even if we were able to build out however much data centers that we want to build out, we still need other forms of power. And so from a startup perspective, I think there's a lot of interesting things happening in the nuclear space. I mean, Panthalos is putting nodes in the ocean. You're seeing more different types of chips being built that require new types of power.

8:26And so I don't think that changes. I think, sure, there's going to be more regulatory rules and regulations around this, but you're still going to see startups continuously be like, OK, there are power constraints for amount of AI compute that's needed in these next two years, five years, 10 years. And so what startups are going to build to be like, hey, maybe we can't build out as much data centers, but maybe we're going to build chips that don't require as much power. And so I'm seeing that a lot happening right now. And I think it's this just shows that that's even more prevalent and more important as we continue to grow and build.

8:59Yeah. Necessity, the mother of all invention. If you can't get power, well, maybe you could use less or maybe you can find new power sources or make the ones you have less wasteful, more efficient. Tomas, who wins in all of this with the split government in the short and the long term? Let's assume the Dems then win in 2028. What does it look like for our industry? Because we had four years of no M &A under the wrath of Lena Kahn. That is top of mind for me. Our industry was just stymied from a lot of acquisitions that made a lot of sense. Now, with the last two years with Trump, everybody's evaluating every possible acquisition.

9:44And we'll talk about some of those acquisitions. Maybe the next two years is the next 18 months, maybe just the next 12 years. 2027 is the last year to get a big deal done. And then you could be looking at five years of no more M &A. Terrible for our industry if that happens.

10:02Tomasz Tunguz: It would be really tough. I mean, we've seen Lena Kahn come out, I think, pro-anthropic. That was an unusual position that she had when it came to protections around civilians, or sorry, citizens, when it comes to what the AI models do. I think we may have six months here in the beginning of 27, because every HSR process or any kind of OGA or the FTC has an 18-month approval process. And many of these acquisitions, like we just saw with the ex-JDS Uniphase business that Marvell bought, or has tried to buy, might be the largest M &A ever in tech history. the long pole in the tent there is actually Chinese approval rather than American approval.

10:40Tomasz Tunguz: So if you really want a big acquisition, you probably have to get through, you have to execute that transaction before June of July or next year, or you're probably within an 18-month window where either the US, the EU, whose posture on acquisitions really hasn't changed, Draghi really has not changed its position on M &A, or China will become an important blocker or at least a constituent in that conversation. All right. Listen, if you're running a company and you're the founder, Hey, you're going to have so much work to do. You got to hire a team. You got to build the product. You got to do customer interviews and understand what your product or service is going to do for your customer.

11:20Well, you're also going to need to incorporate and you're going to have to do your taxes. You have so much work to do on payroll benefits, taxes, banking, incorporation, and you need a partner and that partner is every every has worked with over 1000 startups, you're going to make mistakes. And some of those mistakes are going to be costly. That's why you need a partner like every they're going to handle your Delaware C Corp, your EIN and getting a registered agent with no legal fees and no delays. They're going to get all that done for you. And if you're further along and you need to scale up your team, well, every will help you hire contractors in over 200 countries and do it right.

12:00Tight is right. Plus, their banking platform keeps your funds board compliant while making it easy to collect receipts, pay reimbursements, and stay on top of your spending. So head over to every.io and stop wasting time on stuff that isn't growing your company. That's every.io. It is, to your point, DA, shocking how big of an issue the data centers have become. I don't know why we're here and how we got here exactly, but an NBC exit poll, and it just seems farcical to even say this, but it apparently is true, found 56 % of voters called AI and data centers a crucial or major factor in their vote.

12:44And Democrats spent more than 100 million on ads, tying Republicans to big tech and data centers per time. Doesn't feel like that is matching the reality of what data centers do. Data centers bring a ton of taxes. They don't ask for discounts or tax breaks from local communities anymore. And in fact, they'll guarantee that prices of electricity won't go up. But maybe it's just too late to change this narrative, as you said. Projected winner of the governor of Texas, Gina is not super pro data center. So this seems to be one of the big challenges is turning this narrative around. I think it dovetails with jobs, as you mentioned there, DA.

13:33And something I monitor because I have probably a dozen investments in autonomy, including my original one in Uber, or I still have a significant portion of that. There's been a lot of hand wringing in the Northeast around these self-driving cars, autonomous vehicles, robo taxis, Waymos, Tesla, and then Zoox, AV ride, et cetera. I think there's a non-zero chance that we're going to see a wave of anti-self-driving go across the country. And that's terrible because self-driving saves a lot of lives, but if you're trying to save jobs, that's a place where I don't think you can win the argument, DA, that there's some great benefit to self-driving for employment.

14:22Ultimately, we want every person doing something that is worth their precious life, right? And as soon as you can have a robot drive a car perfectly well, it's sort of a waste of time for a person to do that now of course um realizing that also requires you to have a new job for them to do that's better than the job they were doing and i think there's this interesting tension right now in that you know kind of like the robin hood bros are also uber drivers you know that's what they do at night they do postmates they do uber they do whatever and then they go on robin hood and they trade tesla stock and they trade uber stock and they listen to all in and they right And so it's like navigating all of these conflicting interests among these constituencies is going to be really challenging for politicians.

15:16And I think the backdrop of it all, and you're seeing this play out in Europe in a big way, is that the intergenerational war is the true winter that is coming. Like winter is coming. The Social Security Trust Fund is going to be insolvent in 2032. And the truth is that the people who own equities that are really going to be the beneficiaries of Uber and Tesla and SpaceX and autonomy and data centers are old people. And those old people are not driving Uber. Those old people are chilling in Boca Raton and checking their Schwab account every day and feeling good about how great things are going.

15:58And meanwhile, the 22 year old Uber driver is looking out at a future where their job may be taken. They can't buy a house. It feels like everything's gotten more expensive. And that's why, irrespective of whether they voted for Trump or Kamala last time, they turned out yesterday and they said, no, this isn't working for me. No one is out here actually giving a shit about me and my future. Well said. Intergenerational concerns here. Ilana, and we have the M &A Concerned Jobs Data Centers. This is just going to be a tough setup, I think, for the technology industry. I guess my question is, can the Dems actually find a middle ground here between being anti-capitalist, anti-progress, being decels, and getting economic growth?

16:50Because if they do take this position that young people or socialists want them to take, well, now we're getting into Ayn Rand kind of territory. We're getting into who's going to create these jobs. Where are they going to come from? If you slow things down too much, this could be quite problematic. So your thoughts there on Dems maybe finding a middle ground here? Yeah, a few things there. And I know we were talking before the podcast about, okay, people who can't even drive cars, people who can't afford a car, Well, suddenly now they could get somewhere so much cheaper, maybe even further away for less of the cost than they were before.

17:29And I think like that's a huge opportunity to be like, hey, now they could go somewhere that's further away because they didn't have a car. They're not waiting for a bus that will take them two hours to get to the job site. It will take them 20 minutes by going and getting a roadway taxi. And so I think there are a lot of opportunities there. I don't think our country will go and look at China, look at other kind of countries are building and continue to go and really take in the technology that's being built and be like, no, we're actually just slowing it down here. We'll let you guys have the self-driving cars.

18:00We'll let you guys have the robots. But here in the U.S., we just want to like keep driving our own cars, like not have robots, even though that we have the technology to do so. I think as a country, we're scared of it. Like you always hear this thought of AI is taking over, jobs are done, nobody's going to have anything to do anymore, and there will be no way to make money because AI and robots will be doing it all. And I think with all this, there are more opportunities. When horses stopped and cars came, opportunities came up too. And so I think it's scary. I think we're living in a world where AI agents are doing 50 % of the jobs that were before and all the manual work you no longer need.

18:42But I think that there's still opportunities in there. Who's actually going and building the robots? And you're seeing blue-collar jobs continue to grow and more and more is necessary. And so all in all, I think, like, as a country, we probably need to change our sentiment around technology because we are going to have to embrace it. And we're, as a country, I don't see us just watching other countries continue to succeed and not go and really embrace it here.

19:07Tomasz Tunguz: Jason, you have an amazing perch. Like, who is telling the positive? Like, I remember going to Disneyland and seeing, like, the House of the Future, right? Like, in the 1950s, right? And it looked like the Jetsons. And there was this view that we would be landing rockets and we would have Rosie, who's the household. Like, who? We have, like, movie stars telling this positive story of AI. but have you seen anybody actually with that rosy scenario in america i think if you look at zuckerberg uh as crazy as it seems he created a program for generation tool belt to train people for free to become plumbers electricians etc so he's got such a bad the corporate reputation at meta is the worst of almost every company right um and they have really good comms people there who are like, how do we turn this around?

19:57And that was, I think, one of their efforts. And they started that program. Then they're doing a lot on landing pages about data centers. So you constantly see them putting up these landing pages around, hey, data centers are great. Look at the tax. Look at the short-term jobs, the mid-term jobs, the long-term jobs. We're going to have 4 ,000 people there building this data center in Waco, Texas, or whatever, El Paso. And here's all the downstream effects. Plus, we're going to pay for your kids, you know, new soccer field or whatever it happens to be. And then Muse. I think the thing that's going to turn this around, as crazy as it seems, is when the stay-at-home mom or dad, the teacher, your everyday American starts using an agent.

20:43And the agent starts making their life better. And the agent says, you know what? I went through your emails. You're paying for Netflix twice. Can I unsubscribe from one of those? And they're like, yeah, I don't have time to do that. It's done. And then they say, oh, yeah, and by the way, you have your small plumbing business. I went through all your emails. You missed these two quotes. There's one in your spam filter. And I took the time to write a quote to the person. Would you like me to send it on your behalf? And then Joe the plumber and Susan with her small business, they start actually experiencing gains that we're all experiencing.

21:22I did a tweet earlier this week. I'm spending as a company$40 ,000 on Claude, on GrokBot, Cursor, Lovable, 11-Libs, this long tail of things. And I calculated I'm getting$2 million in value. How did I calculate it? I looked at three different software projects we're doing internally. One to build essentially AngelList, Alana, for the syndicate. We left Dangerless to do our own thing. We were a little bit too big for it. And now that platform, I always wanted to build. It would have taken four people, five people to build it in two years. And we built it in three months with no developers and no designers.

22:00And it works. And it's getting better every day. We did the same thing for our programs and our accelerator. And we did the same thing for the podcast and the media business. Each of those would have been 12 full-time employees that call it 100 to 150K each. Once that trickles down to not a 20-person venture capital firm, but a two-person mom and pop shop, that's like, how do we make more money? And it says, oh yeah, by the way, there's these three other services. I could take your inventory and the photos from your camera roll and put this stuff on eBay for you automatically. And then you'd have two ways to sell from your store.

22:35And do you know about Etsy? I could set up an Etsy store for you and I'll manage the Shopify store for free. And you're currently paying$4 ,000 to some vendor to maintain your Shopify store. I can do that all for free. That's the kind of stuff that's going to make people feel great about it. And if their kids can get a job because their kids embrace the tool and the kid learns how to play guitar or be a plumber or be a ranch hand or be a chef because not just of YouTube, but their agents kind of help them fill in that knowledge. when everybody gets their own personal tutor, when the price of homes goes down significantly, when the price of healthcare goes down because AI is providing it, that's when I think we could turn this around.

23:17But there's nobody leading that charge. Nobody has that narrative.

23:21Tomasz Tunguz: And I think this will happen much faster than we expect. I mean, you look at the Muse numbers, right? You have 10 million people in the first quarter downloading it. Two and a half million people are daily active. You can imagine, okay, let's wake up in 24 months. And now we actually have some meaningful fraction of Americans, 330 million Americans, let's say we have 30 or 40 million people on news every day, all of a sudden we're outstripped of capacity because even now some of these systems are having problems. And then we are in the US a net exporter of tokens, a net importer of tokens. I think one of the narratives that we need to start talking about is just like oil, just like diesel, are we a net importer of tokens or are we a net exporter?

23:58Tomasz Tunguz: And I wonder, you need that positive science fiction view of the world before you can actually have this conversation because there needs to be enough domestic demand and will for us to be producers of tokens to change this data center narrative, which is absolutely essential. Here's a lesson for all you founders out there. Trust closes deals. You can have an amazing, incredible product. It solves every problem your customer has, but if your customers don't believe that it's completely secure and they can trust it, they're out. This is why you need Vanta. They're the number one security and trust platform that's going to get you compliant fast for all the major frameworks from SOC 2 to HIPAA and beyond.

24:36Plus, Vanta is going to make sure you stay compliant by continuously monitoring your controls. And now you can interact with Vanta's agents wherever you're working, even if your team is knee-deep in Claude or Cursor. Having Vanta as a partner means your deals keep moving forward and your engineers can focus on building, not documentation. That's why Vanta is already trusted by more than 16 ,000 companies, including major names like Harvey, Ramp, and Ryder. Prove to your customers that you're ready for business. Learn more at vanta.com slash twist and get$1 ,000 off. That's V-A-N-T-A dot com slash twist for$1 ,000 off.

25:18DA, what do you think? It's an interesting narrative. I think there's the digital domain and then there's the physical real world, right? And if I look at my monthly credit card bill or my bank statement, all the expensive stuff is basically physical stuff. It's my rent, it's healthcare, it's school, it's a babysitter, it's all that kind of stuff. It's a car. And there's this interesting moment we're at right now where it's like AI is here and it can solve a bunch of business problems for business people. But the question is, can it actually solve these problems that are material to an average person's life?

26:04And as I was just hearing you guys talk about this, what occurs to me is like, unfortunately, the real problems that need to be solved in the average person's life are today largely problems that require public policy. And so I actually think the way that AI can be most helpful would be if it helps us as a society to be less partisan and more constructive in our politics. And I don't know that AI had anything to do with the election results yesterday, but just imagine that everyone going to the polling booth is going to be using their AI assistant to help them understand what they're voting on and what would actually be good for them.

26:51I mean, fundamentally, the importance of this technology is that it should make you smarter if you're not that smart, which, you know, I'll be the first to admit I'm not that smart. I'm bad at math in high school. And there's multiple types of intelligence as well. You have artistic, emotional, and pure IQ. You have all three. Well, keep it coming. Keep it coming, Jason. I like that. But here's what I would say. When I get the ballot, okay, it's impossible for me to manage this thing. Okay. It's like a hundred decisions. It's so complicated. I never heard any of these people. I get one sentence on each of them.

27:26So of course, I went on, you know, Gemini in this case, and I just every single one I typed in the candidates names and I said, here's what I care about. Tell me who to vote for. Love it. So imagine that what AI does is it basically is the infrastructure for us to be smarter as a people and stop shooting ourselves in the foot on all of the issues that we need to deal with. Like we have to deal with, you know, permitting issues for construction. You've seen obviously in Texas where you are, how much they've been able to move the needle on this in cities like Austin or Houston. Like it's not as if we don't know how to solve our big problems.

28:11Lurie's doing it in San Francisco. You've got great mayors and governors who are trying things all over the place. That is the beauty of federalism. We get to try things all over the country, figure out what the best ideas are. And then if we were smart, we should just take those best practices and scale them up. And like being partisan is just a huge impediment to doing the smart stuff that we should do. So I think that's my hope. It's like AI helps us be smarter. Being smarter helps us fix our actual problems, which AI can't solve on its own. It takes us to do that. Yeah, you need to have the meat layer between the AI, that software.

28:49Yeah, you love this meat thing. I love the meat thing. I didn't eat breakfast. I'm kind of hungry here, but I need - You're a meat boy. I gotta go get some brisket at the Salt Lake. But in all seriousness, I do think you're onto something with the real world. Another thing that will turn this narrative around is when people start experiencing things like this little tiny robot they attach to their garbage. I love that. Here's an arduous task. And I saw this come across, like a hundred pound trash bin. You gotta drag this thing out. I literally, you know, I have a ranch. So it's literally a half mile from the house to the gate.

29:24There is a service we pay to show up on Sundays, take these from our house, put them in a truck and drive them to the front and then come back the next day and put it back. These like literal companies that do this. Here's a garbage can with a little robotic attachment that Kirby from Revere Robotics. And then when people start experiencing this, we're investors in Zipline. And when people start experiencing their burrito or their extra diapers or their prescription, let's say they got a cold and they need to get a Z-Pak or something, when this stuff starts getting dropped off by Zipline, no delivery fee or the delivery is a buck or it's incorporated into the price is essentially what I think will happen.

30:08There's no tip. And it comes in under 15 minutes, under 10 minutes. This is the kind of stuff that will get people excited about AI and about technology again, Alana, I believe. Yeah, I mean, I have a robot vacuum and I think it's the best thing in the world. My neighbors over here have like a robot lawnmower. And I mean, if a robot could go and be like doing my laundry, folding my clothes, making my bed, like I'm all here for it. But I think people are just like, kind of what we talked about before, just saying, okay, data centers are being built around me. The cost as an individual is going to rise, which just isn't true because there's so many regulations around that right now saying that that isn't the case.

30:45Or people are gonna say, hey, we're gonna lose all our jobs. And I just think what we're all saying here is like, there's also so many positives to this. And there's so many things that are going to come from this, that the manual tasks, the manual labor, maybe people get hurt in driving a car every single day. But we're not really talking about how many lives it saves, because instead of texting and driving and being distracted by driving, we have a robot driving you. And I think the headlines that read is, oh, this Waymo got into an accident. Or I feel like there was a video the other week that like, oh, Waymo made a left turn when it shouldn't.

31:18And like one Waymo in the whole fleet of them made that mistake. They did kill one cat. They killed a cat. But I'm not a cat guy. I'm not a cat girl either. I wasn't particularly broken hearted about them. I'm actually kind of not a fan of cats. I don't like them. They might make that the headline later. Like Waymo kills cat. Jason doesn't care. FSD. I mean, I don't know. Does everybody here drive a Tesla with FSD? I do. I don't, but I'm considering getting a Tesla. What about you, DA? Are you in it? I do. I used to. I don't. I'm waiting for the next version. The next version. Tomas? I hear it's amazing.

31:56Tomasz Tunguz: It's a game changer. I mean, it's a Waymo in your driveway. It's crazy, right? Yeah, it just drives itself. It saved me from two or three accidents. I was driving home in the rain last night in San Francisco, and the reaction times. I think one, I think if you're an insurer, like I wonder if actually the mandate to shift to self-driving will not come from insurance because the - They'll have no business. Yeah, there's just, well, yeah. And so maybe you see the convergence of the self-driving car company with the insurance business, but you just look at like the, I think maybe car insurance ultimately ends up looking like title insurance.

32:31Tomasz Tunguz: We all buy title insurance, we buy a house. The redemption rates on those policies These are less than 2%. And what happens if car insurance, you only get only 2 % or 1 % of people actually who drive in a year actually get into accidents. We can get there, right? Waymo's at 93 % or 95%. So I think it's pretty fundamental. And it's almost, there's this joke on Twitter, right? The most expensive Tesla, the most expensive car I ever bought was the Tesla full self-driving because my spouse wanted one immediately. And so I had to buy two. Yeah. I think you're, you know, there's almost a certain level of - It's a true joke too.

33:06Tomasz Tunguz: Yeah. A certain level of personal responsibility. Like I care about you and my family. I want you to have the safest driving experience. If you have a 16 year old, what are you going to put them in? If you can, you'll put them in a self-driving car so they can't hurt themselves or other people. I think being anti-AI will become increasingly a losing position as consumers get huge value from it. It'd be like being against. J.K.L. Yeah. I got a question for the group, which is, you know, it's like AI is already quite cheap for the end consumer. I mean, you just talked about how disproportionate the value you've gotten in your businesses relative to what it's cost you, but it still could be cheaper.

33:50It could be free. And you think about how China's going at this. As I understand it, their national strategy is basically focus everything on getting AI out there into the economy, because the real beneficiaries are going to be the people like you who are using it in their business, making their businesses better, or individuals making their lives better. And what I wonder is like, there's this thesis in Silicon Valley, the so-called everyone wins thesis. And that is, just to recap it, that the memory makers can win. NVIDIA can win. The Neo clouds can win. The frontier labs can win. Everyone can win.

34:30That is basically an argument that there is so much profit to go around. Yes. That everyone in the ecosystem has this opportunity to basically extract monopoly rents. and what i wonder is like is that sustainable like i don't know that that is at all part of the political conversation around ai yet but i do have to say like it makes me a little uneasy to watch all of these companies making 80 gross margins yeah and that's not how capitalism is supposed to work when you're when you got those fat margins it's supposed to attract competition The margins get pushed down. Your competitive advantage period wanes.

35:15What's going on here? And, you know, we alluded to Lena Khan earlier. Like, is there anything that needs to happen here to either spur more competition or otherwise bring down these margins, which just seem unwarranted? Yeah, this is a key question. And I think what we saw from Canva just this week, Canva cut 22 % of their staff. That is 100 % AI taking and compressing down things people are paying for. How many people do you know who pay for Canva who are moms and dads at home and using it? You know, 1 ,200 people. I think it's 20 % of their staff. I can't believe they have 5 ,000 people working at that company or 6 ,000 people working there.

36:07That's a large number. It feels like if Bending Spoons was running Canva, they'd have 400 people, you know, like 10 % or something. So, yeah, I think a lot of this software is going to become free, Alana. and then a lot of this margin is going to get taken away. Don't you think? Yeah. I mean, if your business relies on kind of third-party inference, what happens when a lot of these customers demand lower prices and you have to continue to cut your margins? And I think a lot of these, I mean, Canva's great, but like they're almost a part of their business is this AI wrapper and they're relying on other AI tools and LLMs to be able to create what they want.

36:48they have to be able to cut their costs. And I think it's going to be really interesting because in the future, I think a lot of AI is going to be discussing these economics. And we've sent years talking about what these models are capable of, but now companies are starting to ask what is actually the cost to deploy them at scale. And so I think companies have been writing this, okay, we need to deploy as much AI as possible in all the different features and aspects, But now they really have to hone in on, hey, if they are using AI, how many people are really necessary to build the same thing that they're building today?

37:22And I mean, I think you were just saying it where you're spending 40K to do, what,$2 million worth of work. And I don't think AI could do everything, but I think AI helps a lot of these jobs do things quicker and easier and not as many people. Tomáš, these were software projects internally that I wanted to do. I tried to do one of them, blew through like$100 ,000, and I gave up because I just couldn't make forward motion on it. And I was like, this is going to wind up costing me a million dollars, and I'm going to get $250K worth of value from it. So something has changed dramatically here. What's your take?

37:54Tomasz Tunguz: Yeah, I'll take the other side, which is, I mean, you've seen GPU costs go up 120 % from about$5 to about$13 in the last six months. I think as a result of the election - That's the hourly cost of a GPU you're talking to. Yeah, for Blackwell 300, which is state of the art today. I think we're going to get to a place where we won't have enough AI in 27. And I think for Muse, we won't have it. For other products, we won't have it. And so we will start to buy AI the way that we buy internet. When you buy internet, you buy a certain amount of bandwidth. How much stuff can I download? And then I buy it as a function of speed.

38:32Tomasz Tunguz: How quickly does it come? We are going to start to buy AI in the same way, which is like how many questions can I ask and how quick is the response? It's going to be rationed is what you're saying. You're going to get an allotment. It's going to be rationed. And this all-you-can-eat stuff is maybe going to go away. Right. And so that goes back to the Canva point, which is, sure, today when inference is cheap and you can get as much of it as you want, vibe code whatever I like. But in an environment where GPU and AI is constrained, not everybody will be able to afford state-of-the-art AI. not everybody will be able to afford Soda performance.

39:06Tomasz Tunguz: And so instead of spending$100 ,000 to VibeCode a custom app, I'll spend$30 ,000 and I'll make do with an app that I have to kind of shim and wedge in. And so I think we'll ultimately see a pretty big resurgence at the app layer as a result, just because of the supply demand will make application, like bespoke application development and maintenance too expensive for most people. So do you think we're not gonna find like a cheaper GPU and a cheaper chip? Because right now, obviously, NVIDIA owns most of the chips or a cheaper way to get power. Yeah, I mean, we're seeing a doubling in intelligence per watt year over year.

39:39Tomasz Tunguz: That's what the head of Google data centers has said. You know, we've seen a 400x reduction in overall cost. But the reality is, like, there is a finite supply. Next year, we'll have probably around 81 gigawatts. We have to decide what fraction of that is inference, how much of that are we selling to each other in order to do stuff for the enterprise. And about 60 % of it is for model training. And if we continue to train these models at$300,$400,$500 million a run, And there's just not that much inference to go around, even with the overall efficiency and the next generation chips, as you point out, Alana, like you have the Vera Rubens coming online and they're twice as productive.

40:10Tomasz Tunguz: But the demand is growing way faster than 200 % per year. You know, it's even harder than writing great code, consistently recruiting people who can write good code. And if you're working on a project and you need to hire great engineers, but you don't want to go through a painful, time-consuming hiring process, G2i can help. G2i has a network of more than 8 ,000 pre-vetted software engineers, the creme de la creme, all with more than five years of coding experience. These are the elite developers you need. And these aren't people who hack some automated coding test. The team at G2i puts engineers through live technical interviews to ensure these are folks who really know how to get their hands dirty building stuff.

40:54And the best part for founders, they make the entire process fast and easy. You get a dedicated Slack channel where you communicate directly with the G2i team so that you can share exactly what you need and start getting matched with engineers today. So go right now to g2i.co slash twist and get$1 ,500 off your first hire. That's g2i.co slash twist. When you say the demand is growing so much fat, I mean, this is what all the companies in the space say too, right? And my question is, well, then why don't the prices just go up? I mean, they could raise the price to the point where there is only the amount of demand that they can meet.

41:34Well, they're already an 85 % margin.

41:36Tomasz Tunguz: They're 85 % approximately. I mean, you can look at the Anthropic S1, approximately 81, 82 % gross margin compared to a publicly traded software company of about 71. And we've seen a price war. So you look at what OpenAI did with Luna, they cut prices 50%, then they cut prices 80%. You have open source model that is significantly less expensive, 80 to 85 % less expensive than closed source. You have a lot of deflationary pressures that are going here. The problem is, think about it like a pharmaceutical company. If it takes me half a billion dollars or a billion dollars to make a drug, if I'm a pharma company, I have 17 years to commercialize that with a patent.

42:09Tomasz Tunguz: If I'm a state-of-the-art AI lab, I have 41 days. because in 41 days, someone else is coming out with another soda model, and then all of a sudden, share starts to shift. And so I need to generate a pretty significant and healthy margin in those 41 days to be able to finance the development of the next model. Capitalism is so brutal, D.A. It's so brutal. I would love to get your speculation, Tomas, about what the steady state margin ought to be for these companies. What you just said sounds like an argument for Frontier Labs having no sustainable margin in steady state. Tell me, one, is that a correct representation?

42:46But two, if you start to think through these other layers, are the memory guys going to be doing 80 % gross margins four years from now, five years from now? I mean, Micron's trading at a forward P of four and a half. Hynix trading at a forward P of 2.8. I mean, if you believe what you're saying about the scale of this, how do you not put all of your retirement funds into Micron stock? I'll be right back. I'm trading like a lunatic. Like, as the market came down, I just bought a million dollars in Uber and Tesla, and I totally hit the bottom. But Tomas, answer the question here about this steady state, because we're kind of evoking Jevon's paradox here, I think.

43:25Tomasz Tunguz: Yeah, I mean, look, I think the problem with the hardware companies and why I wouldn't load up on those stocks is primarily because their GPUs lying fallow in warehouses. We don't have enough electricity to light them up. So at some point, NVIDIA bookings plateau just because we can't deploy them within the U.S. and then you have a seismic shock to those stocks. I think on the other question, DA, like what is the terminal margin for some of these businesses? You look at AWS, right? AWS, massive, the largest hyperscaler. A lot of the business before the AI wave was selling commodity compute. If you rent a server on AWS, it's a server just like there is on Amazon, sorry, on Azure, just the way there is on GCP.

44:03Tomasz Tunguz: The operating margins, not the gross margins, but the operating margins of that business, in other words, the profitability of that particular business unit is 35 to 36%. net. So even if they're selling a commodity good, even if you're an oil company and you're selling crude or diesel, you can still have a really nice, healthy business with competition. Like Coca-Cola's gross margins and net margins are not zero because they're competing with Pepsi. They just continuously try to develop new products and innovate. So I think, you know, having even for like the broader macro economy geopolitically, having these companies generate 70 to 80 % gross margins is very healthy, sustainable.

44:39Tomasz Tunguz: And it's probably the right thing because we want to encourage them to continue to develop and innovate. Yeah. At the$50 to one ratio, I put my spend, Alana, at$50 in value for every dollar. Now I might not have built that stuff. So maybe it's really 10 to one or something. But I see the revenue going up in the projects I'm doing. I'm seeing the ambition, the dollars invested. Everything's going up and to the right because when we deploy the software, we kind of get more efficient. That argues to me that they're undercharging me. And I think everybody's getting undercharged here massively. And plus, I just ordered a$12 ,000 Mac Studio to run local models, which is exactly what I paid for my PC Junior, PC XT, PC AT in the 80s.

45:25They were$4 ,000 computers,$5 ,000 computers, maybe you can get down as low as I think$2 ,000 back then. In today's dollars, that would be three or four X. So here we are, like it used to cost 15, the equivalent of$15 ,000 to buy a desktop computer. You just go back to$15 ,000 desktop computers and you got unlimited, unmetered, well not unlimited, but unmetered local compute. And then just to add to that, like you're still hiring at like your podcast. So it's not like you're doing everything. It's all automated and you're completely done. Like I saw on your Twitter, you're hiring for 10 jobs, 14.

46:01I had to check the numbers. And it's like, okay, sure, you're able to automate a lot of the manual tasks, maybe some of the editing that's made for your podcast, some of the preparation. But there still needs to be hands-on. And I think some people forget that, especially as we continue to grow.

46:15Tomasz Tunguz: But on this local inference point, Jason, I think you're right. In an era or a year in 27 when GPUs are limited, local inference will become really big. We're investors in a company called Olama that has about 10 million people using local inference. Of course. Stanford released a study that 90 % of the work that non-coders do with AI can be executed on a local model. So I think this is a really big wave. And you'll start to do stuff locally. And then when you need a cloud model, you'll spike. And this is one of the ways that we'll maintain. I would have no problem, DA, buying everybody in the company a Mac Studio or NVIDIA, Windows, Spark-enabled computer.

46:50I would spend 20K per person because if I'm spending on average right now, 2K per person per year, it's going to kind of net out eventually if I sell them in five years for half the price. I agree. And I think it's like, as they used to say in the internet 1.0 era, information wants to be free. The internet wants to be free. I think intelligence wants to be free. Yes. And hardware can't be free. And so I don't know how long this takes, but ultimately, you're going to have to, to Tomáš's point, you're going to have to have hardware companies that earn a big enough profit to attract the people who know how to do these crazy, complicated things.

47:38Right now, those margins, maybe people are over-earning. You know, you've got a situation right now where Elon is financially incented to put literal rocket engineers on general contracting type jobs, right? Because it's just so valuable to build these data centers as quickly as possible. That isn't steady state. I mean, it's not what we want those rocket scientists doing over the long run. But still, someone's going to have to be assembling those ASML machines. means someone's going to be having to produce all the stuff that we need to make this um to make this work over the long run and and that i think is the most durable business in a way yeah elon web services came out of nowhere spacex all of a sudden has colossus 2 going four times bigger than colossus one and who better to build real stuff in the real world like factories and data centers and elon he's got really great engineers working for him but yeah they're they're they're shifting key talent, I think, over to Elon Web Services because it's such a money printing business.

48:42When we get back from this quick break, I want to talk about changes at Sequoia, my alma mater, where I started and I was the first scout. The scout program has now been retired at Sequoia. I have feelings about it. Before we do that, I want to just give a shout out to my friends at Fintrix, F-I-N-T-R-X. This is a great subscription that applies to all of us in the venture business. We started a forward motion. We wanted to meet a bunch of family offices. We looked for a database of them. Didn't exist. Then I tweeted, hey, is there a database of this? And the founder of Fintrix contacted me and we got this incredible database.

49:19They are tracking tens of thousands of decision makers across the four or 5 ,000 family offices there are. I started doing late stage deals, Ilana, from my syndicate. We did Zipline. We did Vast Space. We did Adams, Travis's new company. But I needed bigger check sizes, not the average check size from, you know, angel-less level syndicates that I was doing early stage. Those were$5 ,000 to$50 ,000 checks. I needed$250 ,000 to$2 million checks for those allocations. Got on Fintrix, built a database, built some AI, queued everybody up, started doing family office meetings, myself personally and with my team, and it worked.

50:01Go to Fintrix.com to learn more. An incredible database if you want to meet family offices. And the detail they had in there was crazy because I found my family office in there, which was like a family office of two, me and my wife, which we just started allocating stuff in a little more of a thoughtful way. And they had details of us. So thanks to Fintrix for supporting the show and supporting my syndicate, syndicate.com. All right, let's talk about Sequoia. And I'm obviously close to this. Rulof Botha, I met him. He invested in my company. He's no longer there. and a number of the projects he set in motion, along with the rest of the team there, now that he's not leading it anymore, have been retired.

50:43Sequoia published a long post this morning, the Sequoia of Tomorrow, making one year since its generational transition that occurred. The firm is closing its London office, which Sequoia says means several people will be leaving. The firm says this is not about the team. It's sunsetting its batch accelerator arc, which a lot of people really love. They were doing their own high-end bespoke version of Y Combinator, sort of a couple hundred people, a dozen companies, I believe. And the Scouts program, which famously myself and Sam Maltman started our investment careers in, me doing Uber, him doing Stripe in that very early stage in one of their most profitable programs.

51:23And they also, they're shutting that down. And they've removed much of the complexity associated with the Sequoia Capital Fund, which is how they held shares post-IPOs for their LPs if they chose to do that. And they got rid of early and growth stages, which were two different groups there that would focus on those different stages. And now there's back to only one Sequoia. Doug Leoni, who's been super invigorated, a friend of mine, friend of the program, is back in the seat. So they have him making investments and driving the firm. and they've raised$10 billion in new funds. They also got rid of the 40-year-old Monday morning partner meeting cycle.

52:11And they're going to go 24-7 like their founders. Tomas, what do you think of these changes? You know people over there. You know the firm and its history. When you look at these changes, is it just the standard thing that happens when you have a change in management? People, you know, have their projects they want to work on? Or do you think it's a streamlining that has something to do with maybe Andreessen Horowitz, which is also, that's a bunch of different fiefdoms. But what do you think about what's going on at Sequoia?

52:42Tomasz Tunguz: Yeah, I mean, obviously, you know, legendary firm clearly have terrific people leading the operation. I think a couple of different, the things that stood out to me, one was the unification of the early and the growth. And, okay, why is this? Well, it's because if you call a seed stage company a seed stage company, the odds are that it's a$5 million investment, a$50 million investment, or a$500 million investment are roughly equal. This is a bit tongue in cheek, but the idea that we had t-shirt size financing rounds 10 years ago, now any one of those can be a seed. And so the structure of an early fund versus a growth fund no longer really applies in this ecosystem because they have a team of wonderful PhDs or researchers coming from a neolab, they might need$300 million to$500 million just to prove the idea the same way that, say, a biotech would in order to reach that first milestone.

53:34Tomasz Tunguz: And so you need really big funds in order to supply them, hence the unification of those two funds. I think that's a really smart move. Alana, you have a thought there about what's going on at Sequoia and the emergence of these sort of mega funds. Sequoia fought it off as long as they could. They wanted to stay small, bespoke, stick to their knitting, but they had to raise growth funds. They had to do bigger things with other players. Yeah, Menlo, Andreessen Horowitz, et cetera, going big. If everybody's going big, you kind of get left out, right? And I think over the past kind of five years, a lot of these venture funds have become incredibly complicated businesses.

54:14They have their accelerator, they have their scout fund, like kind of everything that you just discussed. And I think at the end of the day, if you look back at their returns, like their job is to find exceptional founders and then continue to invest in them over the course of the C, the series A, the series B, et cetera. And so I think what they're saying is first off, like you still have to continue to make those decisions. Even if you invest in a company at the series B company, it could still make generational returns. If you invest in SpaceX at the series B price, your fund would have done great.

54:41And so I think by combining that all, they're saying, Hey, those decisions are just as important as the early stage companies that are the later because they'll just write a bigger check for the same ownership. I think the one part that was really interesting about this is shutting off their scout program. And you probably have a lot of thoughts here because you're like, hey, me and Sam found great companies. Why would you ever kind of take this away? Because there's some people who don't necessarily want to do VC full time, or maybe they do in this as their step in, or there's some founders in your network that could really find these.

55:10So why take away those specific people that you have early on? Because what do you really get from doing that? You just had people going out and finding the best companies and if they were great, if they didn't, I'd be super curious to know the numbers that came from this. I mean, they must've done a lot of research to be like, hey, we have these scouts. Maybe they weren't the best companies in our portfolio. So like, why are we doing this? And I think there had to be something that came from this decision that ultimately led them to decide not to do it. I think it's complexity. It's obviously sad for me because I was literally the first one.

55:42And Ruloff called me and said, we have this idea for a program. We want you to be the first scout. And then it got to hundreds of them. It was their way of kind of recognizing there's so many companies out there. We can't possibly meet with them all. We can't possibly process them all. And there was a signaling issue in the early days of venture capital where if you did the seed or you put but 100 or 250K, and we were putting 25 and 50K checks in originally, then it went to 100, 250. If you didn't do the next round, then you basically damage the company by not doing the next round. Because everybody would say, well, why is Sequoia not doing it?

56:20They're the best in the business. So you're the loser in the portfolio. So this is a way for them to say, we have no control over the scouts program. If Sam Altman wants to write a check, if Jason wants to write a check, they write it. We don't. All they have to do is submit it. We fund it. So it was a way to get around that originally and spread a wider net. And everybody copied it. So now everybody's got scouts. So the originator of it is getting rid of it. Maybe they just think we want to be simpler and we have such a great reputation. Everybody's going to come to us at some point. And to Tomas's point, and your point as well, Alana, we don't have to get them in the beginning.

56:58We can get them in the middle. We can get them in the end. If it's truly a generational company, we just have to get in there at some point. So we just need really great partners, a really great reputation. and we could miss it. I don't think Sequoia, given that their brand and now how well they've done over their last few funds is having any issue getting in touch with the founders. So I think that also takes into account where they're like, hey, maybe we don't need these smaller guys getting it. And we could just do it from all the people who work here full time and we'll just continue to invest in the best companies.

57:27I do think they should have kept going on ARC and they should have expanded that program. A16Z DA is being incredibly aggressive with their speedrun program. And they're accepting people to speedrun in week 10, or just like we had one company that I think got accepted like two weeks before the demo day. And I think they just have really good terms. They put in 500K at 5 million. Then they have the option to put another 500K in at 20 million. So they just have this way to get to 10 or 12 % of the company with the super pro rata option, which they copied from me. And my common denominator copied from me.

58:02I I put that into our early documents with our accelerator. But it's hard work at the beginning. And then with these programs, they're really like kind of, I don't want to say cut you off, but you kind of have to cut people off or else you'll wind up having thousands of companies that you have a small percentage ownership in that expect you to perform for them like you have a board seat. And that just becomes very untenable for these programs. So that's why you create a program. It's like, hey, we're going to make a quick check and then move on. And I don't think many people want to work in the earliest stages.

58:33I don't think people want to do year one and year zero work because the late stage is so profitable, DA, which is why I popped up the barbell strategy with the syndicate. My syndicate members kept saying, can we get into Zipline? Can we get into Travis's new company? So I said, OK, I'm not going to just fight it on principle. That's what the members want. Let me go ask people. So I asked TK, can I get an allocation? Zipline, if we can get an allocation. Vass Spass, can we get an allocation? And everybody says yes. In fact, out of six companies, everybody said yes. The only one I wasn't able to get an allocation in was Waymo because they were massively oversubscribed and Google wanted all of it.

59:08Or we would have done an allocation there. Well, I think you get it. I mean, let's put this in the context of what you said last week here on the pod, which was that the next wave of unicorns are allocorns. And I love this, right? Can you suck with me? Before I say my thing, I want you to just regurgitate what an allocorn is. So two or three people can get a company to a million dollars. Forget about ramen financing. They can just do it on the weekend. You have people doing this on a regular basis now, building a product with AI, getting$10 ,000 to$50 ,000 to$100 ,000 a month in revenue, and then coming to us and saying, hey, I got this business.

59:46Is it venture ready? And we're like, yes. Yes, it is venture ready. You're at a half million in revenue. So I believe unicorns, obviously, are billion-dollar valuation companies. And Alicorn has wings. And they fly over funding rounds. Many of the companies I'm seeing right now don't want to raise money. They're like, we're profitable. Why would we raise money? And I say, oh, rainy day fund. Maybe you want to be able to buy something. What if the market has a pullback and you become the acquirer as opposed to the person who gets acquired because you run out of money? And they're like, but we're profitable.

1:00:19It's a pretty great argument. So it's going to be harder for venture firms to get a good price and have a decent entry price if founders believe, hey, we don't need you. We don't need your money. And so then that argues to get in early or get in late and be essentially like a crossover hedge fund. Right. Well, I think this is, again, this is like the great tension of this moment. Probably the best businesses that are going to get started in the next several years are going to be started by these incredibly capital efficient young hustlers who are AI native. and the investment opportunity to back them is probably going to be small because they're alicorns.

1:01:05They want to fly over the rounds. They don't want a VC on their board. They don't want to go through this gauntlet that has been standard for the past 20 years. And what I see is that they're financing their companies through their social network. I mean, they don't want to go to some big fund. Like, why would you go to Sequoia unless you really need a lot of money? And to me, everything about that press release is really smart with the possible exception of the$10 billion number. I'm not saying they're doing something dumb for their business, but you have to ask, what is the product of a venture capital firm as far as its LPs go?

1:01:47And what it has meant historically was that it was a high dispersion of returns, but the really good venture funds we're going to deliver just these unimaginably high rates of return. Very, very hard to do that with$10 billion, no matter what you're doing right, no matter how smart you are. So that's the challenge. We're in this moment right now where as these intergenerational transitions are taking place at all these firms, the founders want to monetize in many cases by taking their firms public. And that's what you've seen with Blackstone and Aries and Apollo and all these other big asset managers and other asset classes.

1:02:28Ackman is public, I think, Persian Square. Ackman took his thing public in a closed-end structure. Now, I think you're going to see Andreessen try to go public. And you may see other traditionally venture funds trying to go public because that's how you monetize the equity that you've built up as the last generation founder of one of these things. And that just fundamentally changes the nature of these businesses. I think, ironically, what it creates is an emerging opportunity for small, agile funds who are just playing the old venture game, small funds deeply in the social networks of the most compelling young founders in the world.

1:03:09And they're the ones who are going to have a really good shot at getting this sort of alicorn class. Tomash, sounds like you're a fund. Sounds like what you're doing.

1:03:18Tomasz Tunguz: I hope that's what it is. Yeah, we, I mean, we're, you know, we're only about three years old. We've raised a little less than$800 million. And we don't invest in very many companies. We put about 10 companies per fund. And then we have a team that's actually forward deployed. That's great experience. It goes and builds the go-to-market functions of the companies. So it's not necessarily these businesses that reach profitability. I don't think we have a lot of conversations around how quickly a company can get to profitability. I think on the B2C side, so selling to consumers or product-led growth where you can grow by yourself, there's no doubt that those businesses will be incredibly capital efficient and can finance their own growth.

1:03:55Tomasz Tunguz: But a lot of software, even today, because of the way that we're sort of educating people about how to modernize their businesses, requires lots of increasingly expensive people. I wonder how much of what we're experiencing is the tension between public markets, classic private markets, and this new weird thing in the middle, which is the trading of secondary. And so if we're sitting here, Ilana, and you're doing a syndicate, sometimes are you buying previous people's shares or are you just doing primary? I had at my previous syndicate. We did Anthropic and or all SpaceX. At my current one, we've been going direct, not really looking at secondaries too much.

1:04:36It's an interesting market right now. I mean, people are paying crazy prices. I think the really interesting thing I'm seeing is to get access to the next round funding. So many founders are like, oh, just do an uncop with a 10 % discount, a 15 % discount. And then you're like, okay, they just raised the round. It gets two weeks after it closed. You have no idea when they're going to raise their next round of funding. If you're etched or in stink, maybe it will be like three days from your last round. But I think it's super interesting that now you're seeing these on cap saves. And a lot of LPs are taking on this risk saying, hey, can they execute?

1:05:12You have no additional numbers in their last round. And you're disagreeing to get into their future round of funding that you have no idea the price on. And I do think there's some opportunities that you say, yeah, sure, we'll do on cap. It's generational business. We truly believe in it. But seeing it more and more is pretty interesting. I don't know if that will hold up. top of the market and we just don't do them. We don't do deals. We don't have prorata and for some basic information rights. I'm just like, life is short. Like, why would I do that? Um, but it was really interesting. DA is as this middle boiler room kind of before public, after the VCs have done this work, there's just like this boiler room of people who are just trying to trade in and out of the stock and get a percentage.

1:05:53So they're like, unlike us who, we want to be fans of the company. We want to know the founders. We want to be part of the team. And even if we're not working there every day, and there's a great esprit de corps around this concept, we've now inserted a boiler room trader mentality to the industry where just, I'll take 10 % of whatever I sell. So you want$10 million in this robotic company, $10 million in SpaceX. Yeah, great. I'll just take a million dollars now, and I'll get the trade and there'll be some fees. And who knows if the SPV is going to be recognized by the company or not. Not my problem if it gets unwound.

1:06:35Yeah, I think it's crazy. And it's, look, on the one hand, you've got the investor, sort of so-called investor side of this, which it occurs to me is almost like freelance investment banking. It's like there's this new industry of freelance investment banking where people's, you know, they're not investors. They're brokers, right? They're unlicensed brokers who charge a fee and they make their money on the trade. And it's an interesting thing that has emerged, I think exactly as you alluded to, to fill this gap between public and private markets. And it really speaks to the failures of the public market system that we have today.

1:07:16I mean, what should have been happening for the past decade is these amazing growth companies should have gone public much earlier. And the average American saver should have had a shot to participate in that growth and see their retirement account go up as Palantir goes up, as SpaceX goes up. But instead, the poor, lowly public markets investor is left with the dregs after all these freelance investment bankers take their cut. And the reason it's been allowed to happen is because the founders have had so much leverage and everyone has been so permissive to enable these founders to have, you know, dual share classes, complete autocratic control of their companies.

1:08:03So it's like, why would the founder go public? They don't have to, they're in control of their business. Money's easy to find anywhere. They can, you know, have all these supplicants who are fee takers on the deal who are doing, you know, freelance investment banking out there wherever. And, and so it's like, you know, and by the way, you know, like my, my, you know, I've got the cleanest dirty shirt, but mine's a little dirty too. I've done a couple of these, right. Everyone has, cause it's been free money. And so it's like, but it's not a good system. Like what we want is shareholder capitalism.

1:08:36Like America needs to innovate. We've got amazing entrepreneurs. The broad public should have a shot at participating in these companies. They should go public earlier. Sequoia doesn't need a$10 billion fund. Let your great winners go public to get all that growth capital. It would be such a better system. I agree. I think what they need to do is get rid of the concept of accredited investors. And this is another thing that Trump was really making great progress on, being so pro-capitalist, that will probably revert under Democrats and be protectionist again. I've been shouting, why can't people who have under$150 ,000 a year salary or$200 ,000 a year salary on the weekends take a course and become a sophisticated accredited investor?

1:09:20If you want to take that course, you want to learn how to put your own money to work, and then you want to start a syndicate and you're not an accredited investor, or you want to buy into these companies, let them join that class. And then let venture capital firms have 10 000 members why can't i have 10 000 people in my venture firm uh and have them put a thousand dollars each right like this could get very interesting uh you could have you know a a much different world if people who are quote unquote poor middle class too stupid to invest in this class well what if they could invest in a venture firm and pay their two and twenty put a thousand, put$10 ,000 to work, do five years of$10 ,000 capital costs,$2 ,000 each.

1:10:03An Uber driver could participate in investing in Uber because they were driving Uber and they saw how popular and how much the customer love was. You're an HR person. You see LinkedIn having an impact. You're able to buy$25 ,000 worth of LinkedIn with your 401k. They should allow anybody to get in here. That would be another way to solve your problem, DA, which is give access earlier to a wider swath of people. And I've thought about starting my next fund offshore. And there are some countries, I wouldn't be able to have American LPs, but there are other countries that allow anybody to participate, or they have a sophisticated investor test.

1:10:43We got to get this done. I talked to the head of the SEC about it. They're working on it. It's so crazy because I I always said you could really go gamble your life savings away, go on debt, but you can't go invest in startups if it's something that you believe in. And if you go, who has a huge voice and you want to go raise publicly, suddenly now you're not just an accredited investor. You have to be a qualified purchaser. Now you need, what,$2.5 million? And so I think it's so crazy, this aspect of like, hey, you could go gamble away your life savings, spend it all on whatever it is that you want to buy, but you can't go invest in startups.

1:11:15Tomasz Tunguz: You could put it on black. Or red. And then just to go back to the original point, think about how different the narrative would be if every American had exposure or could have had exposure to Anthropic at the Series D and beyond. And you could ride it the whole way. And so that data center down the street is actually what's paying for your kid's college. What these asset manager scam artists are going to do, what they're in the process of doing, is they're going to try and foist all of these private vehicles that can hold these private company securities on savers. So the argument is well underway that the average person should be able to hold private equity in their 401k account, should be able to hold private credit in their 401k account.

1:11:59Now, the reason this is happening is because ETFs have commoditized active public equity management. All the money is gone from that. You can't charge 2 % anymore to pick a bunch of great tech stocks in a mutual fund. And so what all these big asset managers have done is they've tried to find new strategies that they can launch where they can still charge. And that has been the private stuff. So I think this is all rotten. I want these companies public earlier. I'm sorry, if you've got a$5 billion company that's doing$800 million in revenue, why can't you be public? What are you scared of? Why can't you disclose your financials?

1:12:43Why do we need to wait so long to see Anthropics Financials?

1:12:46Tomasz Tunguz: Yeah, well, there are two regulatory reasons. I mean, the first, just to put this into perspective, I was working with a company that was about to go public, the financing cost of an IPO, this was 10 years ago, over$25 million on a company generating$100 million in revenue. So just the expenses associated with that transaction compared to a private market transaction, which would have been$500 ,000 or$600 ,000, were inordinately high. I think the second thing that really changed this and really allowed a lot of private companies to remain private is the Jobs Act. The Jobs Act allowed a private company to massively increase the number of shareholders.

1:13:21Tomasz Tunguz: And it used to be there were two reasons why you needed to go public. The first was you needed access to a huge quantum of capital was not available in the private markets. That's clearly changed. And the second was you needed to raise from more and more people. And so at some point you exceeded the shareholder count. So there you need to go public. Both of those have been relaxed. And I think the second order unanticipated effect of both of them has been that the total number of publicly traded companies is significantly less. And the reality is the private markets, you know, you can look at private credit, which has absolutely exploded, right?

1:13:50Tomasz Tunguz: for global private credit,$4 trillion. That comes out of the 2008 legislation because now the banks are no longer allowed to lend, so you need to find lenders in the different private credit markets. But I agree with you. I think there is a transparency that really needs to start to happen because a lot of retail capital is starting to come into privates. Privates are exploding because it's really tough for them to go public for one reason or another or just unattractive. And so we have this new sort of emerging market that has a lot of opacity. We need to figure out how to make it more transparent, but I would argue not regulated in the same ways that we have in the past, because otherwise we'll just recreate the same problems that we were, you know, the same second order consequences that the original regulation produced.

1:14:32Well, and I think we've got, I think there, I think the one possibility that maybe hasn't gotten enough heat yet, and it kind of harkens back to the tokenization stuff in the crypto world, is what you're seeing with Hyperliquid, you know, in particular. Explain what it is, because I don't know if everybody understands what high-tech. Well, it's a platform, a protocol. It's a crypto-ish kind of concept where decentralized market participants are basically able to create, effectively, think of it as synthetic securities, whereby you can bet on anything. You can bet on private company valuations.

1:15:10You can bet on individual company milestones. If it were a pharmaceutical company, you could bet on an individual clinical trial, not just need to buy the entire Pfizer stock in their huge portfolio where the one trial you have an opinion on is not going to move the needle on the overall valuation. And this really marries with what we've seen happen in the prediction markets world. I believe we may be right on the doorstep of this whole conversation we just had becoming totally irrelevant because it's kind of going to supersede. I mean, it conceptually supersedes the regulated securities market entirely just as an end run around the whole thing.

1:15:53It's permissionless. I don't need some SPV hustler to get me into Palantir or rather to to and a role or something. I can just make a bet on it in some weird synthetic market. This is, I think this is Hyperliquid's market for Anthropic. I'm not sure.

1:16:18Tomasz Tunguz: No, it's for, it looks like the USDC market, which is - Oh, it's USDC. Yeah, it's a stable coin backed by a US dollar. I agree with you. I mean, so tokenized assets, which is what you're talking about, can you take existing equities and debt and move them on chain and then trade them 24 hours a day? It's growing faster than AI. It's the fastest growing technology market in the world. It allows people outside the U.S. to buy U.S. stocks. It allows U.S. investors to buy Korean stocks. If you really wanted to get a stock, a share of SK Hynek, it's actually really tough. And you should be able to get exposure.

1:16:47Tomasz Tunguz: The other thing that's really interesting about these markets is there's a new kind of options. So for a lot of people, myself included, options are kind of impenetrable. You have all these Greek letters associated with understanding what they do. But there's a thing called a perpetual option or a perp, which makes it much simpler. And so I can decide like, okay, let's see what's happening in the Brazil election. I want to short that economy. Here's a really great instrument for me to do it. And so, you know, many of the major, the largest banks in the world are pushing extremely aggressively into tokenization, both of public and private assets in equities and credit.

1:17:21Tomasz Tunguz: And it'll be a huge wave. So I totally agree with you. I mean, look at Jason, you got to pull up, pull up the Calci perpetuals markets. People need to see this. I mean, what Tomas is talking about is crazy. And if you pull that up, you're going to see you can bet on anything. And the price ticker, I'm sure you've seen this, Tomas, and Ilana, the price ticker is moving in real time. I mean, this is catnip for a gambler's mind. Well, and then Jason just had the Anthropic pair up. And what's amazing is two or three weeks ahead of an IPO, you have price discovery happening on these exchanges, which means as a retail investor, I can say, oh, I think the right valuation for this business is 1.75.

1:18:02Tomasz Tunguz: or two and I can make a bet. And all of a sudden now I have access in a way that I didn't have before. So I think it's a really exciting phenomenon. Wow. You can put on whatever leverage you want here. Yeah. You see that little purple slider over there. Uh-huh.

1:18:21Okay. You can just put on a bet and you're just voting up or down with up to almost seven times leverage. Wow. And they're doing this in real time. And you're going to have this at least outside of the US on virtually every asset, including any derivative you can make up. So I think there's a good chance that this just eats a lot of the capital market if it were left to its own devices. Of course, regulators are not going to let that happen. This is going to be heavily regulated in the US, but we'll see how much they tighten the screws on it. And if you're not buying the underlying asset and enough capital moves to perpetuals, then that would make the existing assets go down in value if too many people move over here?

1:19:17Or this is just crazy. Not really. People.

1:19:20Tomasz Tunguz: No, it's like, I mean, okay, so the history of the options market is you had farmers, who wanted to buy insurance so they could sell their corn a year from now at a particular price. And you look at the derivatives markets, which are options or futures on top of corn or stocks, those tend to be about five to 10 times the size in terms of total trading volume compared to the underlying. So if there's a billion dollars worth of corn that's sold every day, there's probably five to 10 billion of derivatives that are sold on top. And anyway, so those markets tend to be significantly bigger. And it actually makes the price discovery of the underlying asset much healthier.

1:19:55Tomasz Tunguz: Going back to DA's point about how there's a lot less active trading, except for hedge funds in public markets. And so you should actually see a significantly healthier overall private market as a result of a - These are getting very popular in Korea, I understand. People love to bet with this massive leverage and they have stocks. So you can buy an actual stock that's trading at 10 times. So here, if you bought, this, I guess, is the US Tech 100. If you buy this at 11.8 times leverage, a 1 % move feels like 11.8%. So you're just getting much more action. This is like caffeine to meth, like no stopping at cocaine.

1:20:39You're just going straight to Breaking Bad blue meth. Don't recommend, folks. I don't. I mean, I guess if you just want the stock to move in the next two days, I guess you could. I mean, it's interesting. I'm very good at calling bottoms. Well, we've got all kinds of levered ETFs right now, by the way. I mean, look, this is all sort of a symptom of wealth inequality in a way. I mean, people are just, as Charlie Munger used to say, the world's not run by greed, it's run by envy. And the numbers that people are seeing in terms of how rich these anthropic founders are getting and even the employees who were hired last year, this kind of thing drives a mindset of rabid risk taking.

1:21:26and you know if you're that if you're that postmates driver who's trying to figure out how to escape from that life you know and you like a stock it's really tempting to go buy the 3x levered etf in hynix or whatever um you know what in your mind you go look i i mean if if yeah if i lose the two thousand dollars who cares if i hit i do care so it's like putting on a parlay it's like i need to have like and we're and we've told people you know i mean this is again And going back to the beginning of the politics, the tech guys have told people that they've got three years to make all the money they're ever going to make.

1:22:03And after that, the world ends or you're forever frozen in your place in the economic hierarchy. And so people are just throwing caution to the wind. They're doing crazy stuff. And that's not going to be a good formula when their retirement rolls around. All right. Another amazing episode this week in Venture Capital Roundtable. Where can people find you, Tomas? Where can people find you online if they want to follow you and your missives?

1:22:27Tomasz Tunguz: Oh, TomTungus.com or TheoryVC.com. There you go. And Lana, where can people find you? Maybe even if they were interested, participate in your syndicate. Join your syndicate and see what you're investing in. Yeah, I'm under It's Lana Gold on Twitter and LinkedIn. And then my website that Jason has up right now, TwoRoads.VC, just actually updated it. And then if you press apply at the syndicate, you could directly go and join the syndicate. And I mean, as Jason knows, it's basically you could think of it like an email subscription. If you're an accredited investor, there's no money that's necessary.

1:23:01You go and you get an email for the different deals that we're doing and you decide on a deal by deal basis if you want to invest. So yeah, let me know. Deal by deal. Critical. You get to make the decision how much you invest and which deals you're in. DA, your thoughts? Where can people find you? X.com slash DA Wallach. And this is my firm's website, timebioventures.com. I'm an LP. There it is. And for me, go to thisweekinvc.substack.com. We're working hard. This is why I'm hiring six. I was trying to hire six journalists. Nobody wants to be a journalist. I'm hiring six analysts for our associate training program.

1:23:39I'm looking for people who would be willing to write either about AI, defense tech, energy, robotics, venture capital, or startups. So I need analysts who are willing to write coverage of these companies and deals. If that's something that sounds interesting to you, if you're out of school, we can teach you how to be a great writer. Go to careers.launch.co. Careers.launch.co. As you were saying, I'm hiring a lot of people. you know why we built so much software the system's working we have a flywheel so now i'm like well maybe we should start a newsletter around defense tech and energy and robotics but i don't have somebody who's a human who can look at the research being done by ai call people on the phone check the facts and actually write an interesting narrative worth reading that's what i believe the future is i think da people are going to want to find acoustic humans and listen to them meat yeah it's pretty crazy the future is meat yeah the future is meat we'll see you next time bye bye

From the publisher

This Week In Startups is made possible by:

Every: https://every.io/

Vanta: https://www.vanta.com/twist

G2i: http://G2i.co/twist

Plaud: https://Plaud.ai/twist

Fintrx: https://www.fintrx.com/

Today’s show:

Democrats took the House and Senate, and overnight trading suggested a huge market drop. But in the morning, the markets went another way, and D.A. Wallach thinks he knows why. A Congress that can’t get anything done can’t do anything BAD, like start a war or a tariff fight. The only thing markets hate than gridlock is uncertainty.

On the first post-midterm This Week in VC roundtable, Jason is joined by Wallach (Time BioVentures), Elana Gold (Two Roads Capital), and Tomasz Tunguz (Theory Ventures).

Together, they work out what a split government could mean for tech and venture. Data centers are now a political target for both parties. 2027 might be the last major window for big M&A. Plus, D.A. worries that an intergenerational war could be brewing, between the people who own AI’s upside, and the young workers who fear getting replaced.

Plus thoughts on Cava’s 22% layoffs, Sequoia shutting down its Scouts and Arc program, the SPV “boiler room,” and more.

Guests

D.A. Wallach on X: https://x.com/dawallach

Time BioVentures: https://www.timebioventures.com/

Elana Gold on X: https://x.com/ItsElanaGold

Two Roads Capital: https://tworoads.vc/

Tomasz Tunguz on X: https://x.com/ttunguz

Theory Ventures: https://theoryvc.com/

Relevant Links:

NBC exit poll: 56% say AI and data centers were a crucial or major factor (DCD): https://www.datacenterdynamics.com/en/news/nbc-exit-poll-56-say-ai-data-centers-were-a-crucial-or-major-factor-in-congressional-vote-decision/

How AI backlash shaped the 2026 midterms (TIME): https://time.com/article/2026/11/04/midterm-elections-results-ai-industry-data-centers/

Musk on chips outrunning electricity in 2026, with Elon's reply: https://x.com/elonmusk/status/2117649850644705591

Panthalassa raises $140M for wave-powered AI compute at sea (Tom's Hardware): https://tomshardware.com/tech-industry/artificial-intelligence/palantir-co-founder-peter-thiel-backs-usd140m-wave-powered-ai-data-center-startup-panthalassa-aims-to-run-offshore-compute-nodes-using-ocean-energy

Marvell to buy Lumentum in a $115B deal (The Next Web): https://thenextweb.com/news/marvell-lumentum-acquisition-115bn-all-stock

San Francisco mourns bodega cat killed by a Waymo (CBS SF): https://www.cbsnews.com/sanfrancisco/news/sf-mission-district-mourns-store-cat-killed-by-waymo

Meta launches $115M program to train workers for data center jobs (Axios): https://www.axios.com/2026/06/08/meta-data-center-workforce-academy-job-guarantee

Keep track of the value you're getting from AI (TWiST): https://twistartups.substack.com/p/keep-track-of-the-value-youre-getting-da4

Meta's Muse hits 10.9M downloads (Yahoo Finance): https://finance.yahoo.com/markets/article/metas-muse-ai-agent-hits-109-million-downloads-as-chatgpt-slides-25-153532560.html

Curbie (Revere Robotics): https://www.revererobotics.com/

Canva lays off over 1,200 people amid slowing growth (TechCrunch): https://techcrunch.com/2026/11/04/canva-lays-off-over-1200-people-amid-slowing-growth/

Ollama raises $65M as it nears 9M developers (Startup Fortune): https://startupfortune.com/ollama-raises-65-million-as-it-grows-to-nearly-9-million-developers/

Sequoia hits rewind, undoing 20 years of expansion (TechCrunch): https://techcrunch.com/2026/11/04/sequoia-hits-rewind-undoing-20-years-of-expansion/

"The Sequoia of Tomorrow" (Sequoia): https://x.com/sequoia/status/2117961312143855764

Travis Kalanick's Atoms raises $1.7B (TechCrunch): https://techcrunch.com/2026/07/22/travis-kalanicks-robotics-company-raises-1-7b-led-by-a16z/

Vast: https://www.vastspace.com

SEC accredited investor exam could expand private-market access (CNBC): https://www.cnbc.com/2026/10/01/sec-accredited-investor-exam-private-markets.html

General Compute raises a $100M Series A (DCD): https://datacenterdynamics.com/en/news/neocloud-general-compute-raises-100m-in-series-a-funding-round

Hyperliquid CEO Jeff Yan on democratizing pre-IPO access (Motley Fool): https://www.fool.com/investing/2026/10/19/hyperliquid-ceo-jeff-yan-says-perpetual-futures-ma/

Kalshi Perps: https://kalshi.com/perps

Aalo Atomics: https://www.aalo.com

Exowatt: https://www.exowatt.com

Timestamps:

0:00 The Blue Wave: Dems take Congress

0:10 Plaud: If your work depends on conversations - interviews, meetings, calls - you need a Plaud NotePin S. You can check it out at https://plaud.ai/twist and use code TWIST for 10% off!

6:42 What a split gov’t means for defense tech

9:21 Is 2027 the last big window for tech M&A?

11:04 Every.io: For all of your incorporation, banking, payroll, benefits, accounting, taxes or other back-office administration needs, visit https://every.io.

13:44 Are anti-robotaxi laws coming?

18:51 The need for positive AI stories

24:12 Vanta - Get $1000 off your SOC 2 at https://www.vanta.com/twist

25:59 How AI can make voters smarter

31:22 Self-driving and the future of car insurance

40:17 G2i has been recruiting software engineers for over a decade. Go to https://g2i.co/twist to find out how G2i vets engineers, and can help you hire them in days, not weeks.

45:00 Jason’s $12K Mac Studio

50:12 Sequoia retires its Scouts and Arc programs

58:21 Jason’s barbell: Zipline, Atoms, and Vast

1:01:19 When VC firms go public

1:08:42 Scrap the accredited investor rule

1:14:26 Hyperliquid and permissionless synthetic markets

1:17:17 Kalshi perps and Anthropic price discovery

Subscribe to our newsletters on Substack:

TWiST: https://twistartups.substack.com/

TWiAI: https://www.thisweekinai.ai/

TWiVC: https://thisweekinvc.substack.com/

Subscribe to This Week in Startups on Apple: https://rb.gy/v19fcp

Follow Jason:

X: https://twitter.com/Jason

LinkedIn: https://www.linkedin.com/in/jasoncalacanis

Check out all our partner offers: https://partners.launch.co/

Great TWIST interviews: Will Guidara, Eoghan McCabe, Steve Huffman, Brian Chesky, Bob Moesta, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland

Follow TWiST:

Twitter: https://twitter.com/TWiStartups

YouTube: https://www.youtube.com/thisweekin

Instagram: https://www.instagram.com/thisweekinstartups

TikTok: https://www.tiktok.com/@thisweekinstartups

Substack: https://twistartups.substack.com

More from This Week in Startups

All episodes