Why Chip Supply Shortages Won't Last | Samir Kaul
MTS
November 4, 2026
Khosla Ventures founding partner Samir Kaul joins MTS to discuss why compute is no longer the real bottleneck in AI biology, how the semiconductor supply shortage will break within five years, and why energy constraints will surpass chip limits in the next 12 months.
Samir Kaul
Google has TPUs, and they're selling them to their arch rivals, Amazon and OpenAI have announced they're buying TPUs. If Google didn't have a surplus, if there's really a shortage of supply, they'd be hoarding it themselves, surely not selling it to their competitors.
So to think that that's gonna be the case five years out, it's just bananas.
SPEAKER_2
Everyone, welcome back to MTS today. Samir Kaul, who is the founding partner and managing director at Khosla Ventures, really excited to have you on. Thank you for joining us.
Samir Kaul
I'm psyched to be here. I love the office.
SPEAKER_2
Yeah, it's a cool, cool spot. I know right before we're going live, we're talking about today is an exciting day. You had a fascinating thesis around prediction markets, whether they're...
Samir Kaul
Well, first of all, it's my daughter's 15th birthday, so happy birthday, Nia. I'm really excited to see what happens the next couple of days on the markets because most of these prediction markets, especially if they have a betting component, the way bookmakers effectively make money is they want an even amount of dollars on either side of the bet so it washes itself out and then they collect whatever fees and they have. So, Kalshi, I would guess the users of Kalshi aren't fully indicative of the general population, right? So, we'll see, is there a bias one way or the other if the prediction markets don't correlate to actually what happens?
SPEAKER_2
Yeah.
Samir Kaul
Because you're basically changing the odds and changing the payouts based on the volume of bets that are coming on one side or the other. So, for example, if the Kalshi user base is more Republican, for example, they'd be more skewed to bet on the Republicans, and so the odds that would be skewed in one way, and the same if they were more skewed toward Democrats.
SPEAKER_2
Yeah. This is going to be pretty interesting because right now, Kalshi has the most users that it's ever had during an election, last election, of course, had prediction markets. It was really popular, but it didn't have the amount of users that it has maybe today. And so in theory, I guess, as these markets become more efficient, they'll become better predictors of truth, but I don't know.
Samir Kaul
It has to have a user base because they're the ones putting the bets in that are really indicative of the environment. So what will be a really good analysis to do in a week is, did Nate Silver's 538, which has no betting component, do better or was more predictive than Khosla or not? That will be a super interesting thing to look for.
SPEAKER_2
That will be really fascinating. I feel like there's all these different lore bits I want to go down in our conversation, and I want to get to, I think there's a lot of interesting areas around your past investment thesis, and also some of your latest investment thesis, specifically when it comes to genome sequencing and how that's all playing out. Before we get all into that, tell me a little bit about the origin story here, around Khosla Ventures.
Samir Kaul
So, I was at Flagship Ventures, where I was an entrepreneur and a partner, and I was actually on a hike in Napa, and I get a call from a random unknown number, and I answered it, and it's someone with a strong Indian accent saying, hi, this is Vinod Khosla, Samir there. And I thought, why would Vinod even know who I am, and what is this? I thought it was a practical joke, I almost hung up.
And he said, no, no, no, I'm in Boston. I started a company he wanted to invest in in Boston. So I'll fast forward, he invested in three companies. He started, and then in 5 said, look, I'm going to start my own firm, I'd love for you to join. Now he charitably called me a founding partner, I was basically came to work for him. And that was 20, almost 22 years ago. And we started with $100 million of Vinod's money. And we were in the back office of Kleiner Perkins for the first year until they kicked us out. And then we've now grown to north of 30 billion under management and the like.
SPEAKER_2
Man, what pieces of your personal investment thesis and areas that you care about did you bring into Khosla Ventures?
Samir Kaul
Well, I was a scientist by training. I had a bunch of publications in Nature and Science. I led the largest plant genome sequencing project out there. So I had a strong scientific background.
Read the other 19 minutes of this episode here.
$15 gets you this episode and 99 more. Credits never expire. After checkout, one link brings you back here to read it.
Not what you expected? Full refund within 14 days.
Thousands of transcripts fetched by people building searchable podcast archives
Fetch transcripts with code
The same credits work from a script or an AI agent. The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000799722787