Who is on this subjectordered by the date of their first quote here
6 of 10 lanes rest on fewer than 5 quotes and are marked thin. Offsets are days from the middle first-quote date, 13 Oct 2022 — a date, and nothing else. It is not a claim about who reached a view first.
-
Bill Gurley
6
11 Jun 2011 – 21 Jul 2023
-4142 d
-
Marc Rowan
5
15 Jul 2021 – 5 Dec 2023
-455 d
-
Gavin Baker
2
thin
25 Jan 2022 – 25 Jan 2022
-261 d
-
Howard Marks
9
2 May 2022 – 4 May 2026
-164 d
-
David Tepper
3
thin
13 Oct 2022 – 6 Feb 2026
+0 d
- Dylan Patel 12 2 Feb 2023 – 16 Jul 2026 +112 d
-
Jensen Huang
4
thin
13 Oct 2024 – 5 Feb 2026
+731 d
-
David Friedberg
1
thin
3 Oct 2025 – 3 Oct 2025
+1086 d
-
Brad Gerstner
2
thin
7 Jan 2026 – 5 May 2026
+1182 d
-
Chamath Palihapitiya
2
thin
3 Jul 2026 – 14 Jul 2026
+1359 d
The chronologysourced and dated, oldest first
-
-
Bill Gurley
Gurley says Benchmark will have four IPOs this year, unprecedented for a single venture firm in a long time.
“Earlier today, one of our investments, Ambarella, filed, which will be our fourth IPO this year. And that hasn't happened, you know, for a single venture firm in a very long time.”
-
-
-
Bill Gurley
Gurley says going public increases pressure and transparency, elevating companies to new performance levels.
“I one of the reasons I encourage companies going public is because I think it ups their game. I think it it presents more pressure. It forces them to be more disclosive, more transparent.”
-
-
-
Marc Rowan
Rowan cites 44% net returns on Apollo's 2001 fund to counter concerns about too much money chasing deals.
“And in that fund, which was the two thousand and one fund, 44% net returns. So now you fast forward and you look at the last three funds we've done.”
-
Bill Gurley
Gurley reports seven of eight direct listings trade above first match, outperforming traditional IPO cohorts.
“Yeah, there've been eight high profile ones. Seven of them interestingly are trading above that first match that they did, which is really astounding.”
-
Bill Gurley
Gurley cites Jay Ritter data showing direct listing companies wildly outperformed traditional IPO peers.
“Jay Ritter put out some data at the end of August that you guys probably saw where he analyzed all the companies that have chosen the direct listing path and they've wildly outperformed their IPO peers, which I think is interesting”
-
Bill Gurley
Gurley cites Jay Ritter's data showing direct listing companies wildly outperformed traditional IPO peers.
“Jay Ritter put out some data at the end of August that you guys probably saw where he analyzed all the companies that have chosen the direct listing path and they've wildly outperformed their IPO peers,”
-
-
-
Gavin Baker
Baker says Nasdaq returns in 2021 were dominated by Google, Microsoft, Nvidia, and Tesla, requiring 60% concentration.
“The market return, particularly for the Nasdaq, was really dominated by a few stocks, Google, Microsoft, Nvidia, Tesla.”
-
Gavin Baker
Baker says great public equity investors are often wrong more than right, with best batting averages around 55%.
“They're wrong more than they're right, but because they make more money when they're right, they still have a great track record.”
-
Marc Rowan
Rowan reports Apollo's thirty-two year track record shows 36% gross and 26% net returns with every fund making carry.
“If you look at the now thirty two year history, 36% gross 26% net returns every fund has made carry.”
-
Howard Marks
Marks argues average returns with below-average risk are equally significant but overlooked because only returns are visible.
“achieving an average return with below average risk is an equally significant accomplish, but easily overlooked because only the risk is evident, only the return is evident.”
-
“Was he was in every year for fourteen years in a row. He was between the twenty seventh and the forty seventh percentile measuring from the top.”
-
Howard Marks
Marks advocates cutting off the bottom tail rather than shooting for the top tail to achieve best long-term performance.
“Approach is very simple, cut off the bottom tail. That's what this guy in the Midwest did.”
-
David Tepper
Tepper notes Panthers started 1-4 this year after winning only five games each of the last two seasons.
“We started off on a four this year. We, you know, had five wins the last two seasons. We, you know, had a different lack of success at other seasons.”
-
-
-
Dylan Patel
Patel says NVIDIA GPU bandwidth increased less than 10x while FLOPS increased 100x from 2016 to 2023.
“The bandwidth has not even gone up one order of magnitude. Right? Whereas flops have gone up two orders of magnitude. So so less than one order of magnitude versus two orders of magnitude increase.”
-
Dylan Patel
Patel reports PyTorch 2.0 delivered 30% performance improvement on 7,000 popular GitHub models.
“So I think I think I think, the PyTorch Foundation tested the 7,000 most popular most starred models on GitHub, and they got, like, a 30% increase.”
-
Dylan Patel
Patel reports PyTorch 2.0 delivered 30 percent performance increase on 7,000 popular models, driving adoption.
“PyTorch Foundation tested the 7,000 most popular most starred models on GitHub, and they got, like, a 30% increase.”
-
Marc Rowan
Rowan says Athene has grown book value at 17% annually for twelve years.
“And when I look at what we've done, we have actually grown book value at about 17% for the last twelve years.”
-
Bill Gurley
Gurley says venture capital must be measured over 20-40 years, longer than most employees stay.
“You got to look at it over twenty to forty years, which is really bizarre because most firms don't have employees that long.”
-
Marc Rowan
Rowan states active management has failed to beat the index 85% of the time for twenty years.
“If you look at the active management, active management has failed to beat the index 85% of the time for twenty years.”
-
Marc Rowan
Rowan reports Apollo's best year ever with around $650 billion in assets under management.
“Single best year in Apollo's history. Earnings. Asset performance. The our platform as you think about it we are around 650,000,000,000 of assets under management in our asset management business.”
-
-
-
Jensen Huang
Huang claims NVIDIA reduced computing costs 100,000x in ten years versus Moore's Law's 100x.
“we drove the marginal cost of computing down by 100000x over the course of ten years. Moore's law would have been about 100x. And we did it it in several ways.”
-
-
-
Dylan Patel
Patel reports reasoning models improved college math test scores from 4% to 80% in six months.
“Basically, our college level math tests were initially 4% for before this innovation was happened to now they're scoring 80%, the best leading edge models. Right? And this has just happened in six months.”
-
Howard Marks
Marks states most equity mutual funds underperform the S&P 500 even before fees.
“if you look at equity mutual funds, most equity mutual funds don't even do as well as the S and P five hundred before fees,”
-
Howard Marks
Marks says beating competitors requires either more winners or fewer losers, rarely both simultaneously.
“I can have more of the things that go up a lot than you do or less of the things that go down a lot than you do.”
-
David Friedberg
Friedberg argues that dwelling on past failures prevents focusing on present opportunities and causes self-harm.
“Because if you don't move on from all of that, you're not gonna focus on what's in front of you.”
-
Dylan Patel
Patel says Blackwell cost is up 50-70% over Hopper but performance exceeds 2x, making it an obvious choice.
“while cost is only up, know, call it 50% or 60% or 70% versus Hopper, the performance is well north of 2x, right?”
-
Dylan Patel
Patel predicts another 50-100% Blackwell performance improvement from software optimization of standard libraries.
“I see another 50% to doubling in the cards, just from the fact that there is so much low hanging fruit or performance left on the table if you are using the general open libraries, right? Whether it's cuDNN, VLM, SGLANG, all these other things,”
-
Dylan Patel
Patel predicts general-purpose Blackwell users could see another performance doubling over the next year from software improvements.
“for the general purpose user, yeah, you're going to have another doubling of performance potentially on the tables over the next year.”
-
Howard Marks
Marks cites math where consistently staying between 27th and 47th percentile for fourteen years produced overall fourth percentile performance.
“So solidly in the second quarter for fourteen years in a row. But interestingly, as a result, for the fourteen years overall, they were in the fourth percentile.”
-
Howard Marks
Marks says General Mills achieved fourth percentile over fourteen years by staying between twenty-seventh and forty-seventh annually.
“for the fourteen years overall, they were in the fourth percentile. Now that's incredible math. You would say, well, if you bounce back in '40 between '27 47, on average, you're probably about 37.”
-
Howard Marks
Marks notes gold returned 7.7% annually since 2010 versus 12.7% for S&P 500 over the same period.
“And if you bought the S and P at the same time, you've had a 12.7% rate of return.”
-
-
-
Brad Gerstner
Gerstner cites 5x inference and 3.5x training performance improvements on NVIDIA Rubin chips.
“He just improved his inference performance by five x, not even including Grok, which I think will give him, you know, another huge boost.”
-
Jensen Huang
Huang contrasts Moore's law (100x per decade) with current AI acceleration (1 million times per decade).
“Remember, Moore's law was two times every 18, 10 times every five years, a 100 times every ten. Okay? But where are we now? A million times every ten years.”
-
Jensen Huang
Huang claims PhysicsNEMO can predict physics simulations ten thousand times faster than traditional methods.
“so it's grounded in the laws of physics, but able to predict 10,000 times faster.”
-
Jensen Huang
Huang claims PhysicsNEMO can predict physics 10,000 times faster than traditional simulation while remaining grounded in physical laws.
“it's grounded in the laws of physics, but able to predict 10,000 times faster. And now, if everything is already running in real time, then you can predict it 10,000 times greater scale.”
-
David Tepper
Tepper claims Appaloosa has the best record over thirty-five years of any hedge fund alive.
“Appaloosa is probably has the best record over the last thirty five years of any hedge fund alive.”
-
David Tepper
Tepper claims Appaloosa has the best record over thirty-five years of any hedge fund alive.
“Appaloosa is probably has the best record over the last thirty five years of any hedge fund alive. That's actually true what it's worth.”
-
Dylan Patel
Patel says networking represents 15-20% of cluster cost but has outsized impact on performance.
“with GPU rental pricing soaring, networking is 1520% of the cost of the cluster, but it can have a much more outsized impact on performance.”
-
Dylan Patel
Patel claims network performance drives 5x to 10x difference in inference for models using disaggregated PD or wide EP techniques.
“And so if you look into inference x, it turns out network performance drives a not just a, hey, 30%. It's actually multiple, like, x's.”
-
Howard Marks
Marks reports Oaktree's high yield default rate is roughly one-third the market average over forty years.
“And so, over the last forty years, on average, something like 3.6 or 3.7% of all high yield bonds have gone into default every year, and our default rate has been roughly a third.”
-
Brad Gerstner
Gerstner reports last month was Altimeter's best month ever in over twenty years.
“It was our best month in the history of Altimeter. And we're bouncing off, you know, a bottom that was a, you know, a tough month the month before, but we stayed the course.”
-
Dylan Patel
Patel claims hardware improved 30x from Hopper to Blackwell for DeepSeek on optimized deployments.
“from Hopper to Blackwell, which is all we've had over the last three years, roughly 30x improvement on DeepSeek, on the most optimized deployment, which you can see on InferenceX there's about a 30x improvement.”
-
Chamath Palihapitiya
Palihapitiya reports their harness with open source models achieved 16.4x cost savings versus Anthropic alone.
“But if you wrap the open source model with our software factory, it was 16.4x cheaper.”
-
Dylan Patel
Patel says SemiAnalysis initially doubted Jensen's 25x Blackwell claim, predicting only 15-20x improvement.
“Jensen, when he originally launched Blackwell, had claimed it would be a 25x improvement. And at the time, no one believed him. Right? It's Jensen, right?”
-
Chamath Palihapitiya
Palihapitiya defends SPAC companies by citing tens of billions raised and millions of customers served.
“The answer is tens billions of dollars. How many clients or patients or customers do they serve? The answer is millions to tens of millions. How have they grown revenues and traction since that happened?”
-
Dylan Patel
Patel reports cache hit rates above 95% for many agentic workflows in production.
“we're seeing cache hit rates above 95% for many AgenTeq workflows, which which means the cost for a cache hit is it's not free,”
-
Where they overlapsame programme, same subject
- CNBC Television also carries Bill Gurley and Chamath Palihapitiya on this subject, at different times.