E26: State of Venture Capital, plus fan questions on longevity, decentralization & quantum computing

Sat, 20 Mar 2021 00:56:56 +0000

Back to episodes
David Friedberg Too Early 00:08:10 healthscience

At least one person who was alive as of March 2021 will live to an age greater than 200 years.

“So arguably, someone is alive on planet Earth today that could live past the age of 200.” View on YouTube
Explanation

No person has yet approached anywhere near 200 years old (the verified human longevity record remains 122 years); this prediction cannot resolve for well over a century.

Chamath Palihapitiya Too Early 00:09:45 healthai

As full‑body MRI providers like Prenuvo accumulate more imaging data and improve their machine‑learning models, the duration of a full‑body MRI scan will shrink from around 60–90 minutes to roughly 40 minutes, and the per-scan price will decline from about $1,500–$2,500 into the low hundreds of dollars, while maintaining equivalent diagnostic resolution.

“And so as a result of that, they're able to actually take the MRI images, and they can do fewer and fewer scans to get the equivalent resolution. So two things will happen. One is the scan time will come down, right? So it was an hour and a half. Now it's sort of right under an hour. It can probably be as low as 40 minutes, which is relatively tolerable for most people. And then the cost will come down into the hundreds of dollars because you'll be able to very rapidly assess.” View on YouTube
Explanation

No independently verified data confirms Prenuvo-style full-body MRI scans have reached roughly 40 minutes and low-hundreds-of-dollars pricing at equivalent resolution.

David Friedberg Too Early 00:25:10 techai

According to the view endorsed by Friedberg, there is under a 5% probability that a quantum computer capable of breaking RSA-2048 via ~4011 logical qubits will exist before 2040, and such a machine is most likely to emerge sometime between 2040 and 2060, at which point RSA-2048–based cryptosystems (including those used for cryptocurrencies) will be practically breakable.

“there is an estimate that the number of logical qubits needed to crack RSA 2048, which is the big kind of encryption standard, um, which could kind of break the whole cryptocurrency model. Um, it would require about 4011 logical qubits.... And by some estimates, and some people have tried to estimate when this would happen. And the estimate currently by some researchers is that there's a less than 5% chance this happens before the year 2040. So we're talking somewhere between the year 2040 and the 2060, when we get a quantum computer that has enough logical qubits to be able to crack a problem like RSA 2048 and basically make all crypto fail.” View on YouTube
Explanation

The predicted resolution window (2040-2060) is more than a decade from starting.

David Friedberg Right 00:26:40 techscience

Between roughly 2021 and 2031, noisy, low-fidelity quantum computers will generate practical breakthroughs in simulating quantum states of materials, atoms, and molecules—leading to advances such as improved enzymes (e.g., for nitrogen fixation, drug discovery, or targeted proteins)—and this application area will be actively and broadly pursued during the 2020s using the quantum compute power already available.

“Now, what's more interesting is that over the next, next ten years or so, the current super noisy, super low fidelity quantum computers can be used to simulate quantum states... That's where the breakthroughs will happen in the next decade.... This decade, we are already going to start to see quantum computers have breakthroughs in how material, how atoms and molecules interact with each other, for example, finding proteins that can do a better job of having an enzymatic reaction in the physical world... And that's going to be chased really hard this decade. Everyone's going to be going after it, and we already have enough compute power.” View on YouTube
Explanation

Near-term noisy quantum computers have indeed been used through the 2020s for genuine progress in simulating molecular and material quantum states, an active research area at Google, IBM, and others contributing to drug discovery and materials science.

David Sacks Too Early 00:57:40 ventureeconomy

Global venture capital annual investment volume will reach approximately $1 trillion by the year 2030.

“Wouldn't be surprised if VC hits 1 trillion by 2030.” View on YouTube
Explanation

Global annual VC investment has been running in the roughly $200-370B range through 2025, well short of $1 trillion, but the predicted 2030 deadline hasn't arrived.

Jason Calacanis Too Early 01:06:48 venture

Over the course of calendar year 2021, the syndicate.com will complete roughly 60 investment deals (about five per month) and deploy a total of about $50 million in capital.

“We're on track this year to do five deals a month, 60 deals, and we'll put 50 million to work this year at the syndicate.com.” View on YouTube
Explanation

No independent data source confirms syndicate.com's specific 2021 deal count or capital deployed.