Who is on this subjectordered by the date of their first quote here
3 of 4 lanes rest on fewer than 5 quotes and are marked thin. Offsets are days from the middle first-quote date, 8 Feb 2024 — a date, and nothing else. It is not a claim about who reached a view first.
- Dylan Patel 3 thin 2 Feb 2023 – 7 Apr 2026 -371 d
-
Bill Gurley
3
thin
8 Feb 2024 – 23 Apr 2024
+0 d
-
Brad Gerstner
3
thin
22 Feb 2024 – 5 May 2026
+14 d
-
Gavin Baker
17
4 Jan 2025 – 4 Aug 2026
+331 d
The chronologysourced and dated, oldest first
-
-
Dylan Patel
Patel states memory cost has quadrupled in NVIDIA GPUs while per-gigabyte pricing remained flat since 2016.
“And now with the h 100, they have 96 or 80 gigabytes of memory, but the cost per gigabyte is the same.”
-
-
-
Bill Gurley
Gurley identifies long-term memory as a major unsolved problem preventing AI from becoming true personal assistants.
“I do think this memory issue is a big, big deal, at least on the consumer side. So none of the major contenders today can remember who you are because it would require them.”
-
Bill Gurley
Gurley says AI leaders admit memory is a hard problem requiring expensive nightly retraining per user.
“If you talk to the people that are at the tops of these firms and you say, hey, why can't this thing remember everything I want? And they go, oh, that's a hard problem.”
-
Brad Gerstner
Gerstner argues memory and context matter more than raw intelligence for AI assistants.
“Or would you like to have an assistant that's just above average intelligence, knows everything about you, and they can use the Internet?”
-
Bill Gurley
Gurley argues initial LLM versions are structurally flawed around personal memory and RAG approaches don't solve it.
“I do think that the initial versions of these models are structurally flawed around providing personal memory and allowing someone to become dependent on one of these things.”
-
-
-
Gavin Baker
Baker reveals high bandwidth memory is a bigger cost component for NVIDIA than Taiwan Semi, with only two capable manufacturers.
“High bandwidth memory is a bigger part of NVIDIA's cogs on GPUs than Taiwan Semi is. And today, there's two companies that can make it, Hynix and Micron.”
-
Gavin Baker
Baker reveals high bandwidth memory is a bigger cost component for NVIDIA than Taiwan Semi, with only two capable manufacturers.
“High bandwidth memory is a bigger part of NVIDIA's cogs on GPUs than Taiwan Semi is. And today, there's two companies that can make it, Hynix and Micron.”
-
Dylan Patel
Patel explains reasoning models dramatically increase serving costs due to long output context and memory constraints.
“your memory usage is going way up with these reasoning models, and you still have a lot of users. So effectively, the cost to serve multiplies by a ton.”
-
Gavin Baker
Baker warns a true DRAM capacity cycle could see prices increase 10X, not the typical 30-50%.
“If the price of DRAM, if like a DRAM wafer is valued at like a five carat diamond.”
-
-
-
Dylan Patel
Patel reports memory prices have quadrupled over the past year, with SSDs up three to four times.
“Memory has like four x in price over the last like year, and it's gonna continue to go up. And then now storage SSD, right?”
-
Brad Gerstner
Gerstner notes memory companies trade at 5x fully taxed earnings and Samsung will profit more than Google this year.
“They're they're trading at five x fully taxed earnings. Samsung's gonna do more in profits this year than Google.”
-
Brad Gerstner
Gerstner says memory trades at 5x earnings and NVIDIA at 13-14x fully taxed GAAP earnings.
“Think about memory as an industry is only trading at five times earning. In the case of NVIDIA, trading at 13 or 14, fully taxed gap earnings.”
-
Gavin Baker
Baker admits he has been hanging on to memory stocks for dear life despite contrarian instincts.
“I've been hanging on to the memory stocks for dear life. And yeah, but that's that's a that that's been a lifelong journey”
-
Gavin Baker
Baker admits he has been hanging on to memory stocks for dear life despite contrarian instincts.
“I've been hanging on to the memory stocks for dear life. And yeah, but that's that's a that that's been a lifelong journey”
-
“I would say based on every memory cycle we have had for the last twenty five years, this is the time to be selling memory 100%.”
-
“I would say based on every memory cycle we have had for the last twenty five years, this is the time to be selling memory 100%.”
-
Gavin Baker
Baker takes the over on every AI capex and memory forecast number given by other speakers at the conference.
“And I would just say, I take the over on every number that they gave. Every single number.”
-
Gavin Baker
Baker argues this may be the first true capacity cycle and that fundamental shortages help avoid a destructive bubble.
“This may be the first true capacity cycle. And I and and I and I do think that these fundamental shortages are good for us as investors.”
-
Gavin Baker
Baker believes NVIDIA could sell $11.52 trillion in chips next year if TSMC tripled capacity but bubble risks prevent it.
“And if they doubled or tripled capacity, like, NVIDIA could probably sell $11,520,000,000,000 worth of chips next year. I really believe that. But the other side of that might be very painful for everyone.”
-
Gavin Baker
Baker predicts GPUs will have 10-15 year useful lives due to prefill-inference disaggregation, extending older chips' value.
“The disaggregation of inference means that I think these GPUs are going to have ten or fifteen year lives.”
-
Gavin Baker
Baker identifies memory as the single most important factor for increasing token output per compute unit.
“It's the single most important thing you could do to increase token output per unit of compute, and then that obviously, definitionally, actually lowers costs,”
-
Gavin Baker
Baker explains breaking memory LTAs could destroy companies by losing allocations when market tightens again.
“Well, if they're breaking their LTAs, it probably means, you know, your oversupply, prices are coming down, and then, you know, capacity naturally contracts. Well, what do you think is gonna happen to Google's allocations?”
-
Gavin Baker
Baker suggests memory companies should copy NVIDIA's credit wrapper model to participate in GPU revenue streams as credit markets revolt.
“Which is? I would be going to the buyers of GPUs, tradiums, and whoever and say, I'll participate in the NVIDIA credit wrapper.”
-
Gavin Baker
Baker suggests memory companies should copy NVIDIA's credit wrapper model while they have cash and credit markets are tight.
“I'm just making this up. But, like, do something. Like, you can because you have money now and credit markets are revolting.”
-
Gavin Baker
Baker argues breaking memory supply agreements could destroy franchises due to allocation loss risk.
“I just think given that this is, the axis around which kind of everything is revolving, man, like, might blow up your entire business and your franchise by breaking an LTA.”
-
Gavin Baker
Baker suggests memory companies should copy NVIDIA's credit wrapper model despite business instability risks.
“I would be going to the buyers of GPUs, Traniums, and whoever and say, I'll participate in the NVIDIA credit wrapper. Now their business is just inherently less stable and predictable,”
-
Where they overlapsame programme, same subject
- Bill Gurley and Brad Gerstner — BG2 Pod, 1 appearance together
- Invest Like the Best also carries Bill Gurley and Gavin Baker on this subject, at different times.