← pool

🤖 We run out of compute again in 2025

O3-style test-time compute plus reasoning models mean GPUs/accelerators sell out in 2025 the way they did in 2023 — inference compute is the derivative winner; 'long NVIDIA and chip makers.'

0 CONVICTION
peaked 67.0 GREEN
EMBER
band
▲ LONG NVDA
expression · bullish tech
HIT
outcome · R +26.4% · α +10.3%
2026-01-04
window closed
⚠ CONFLICTING IDEA ON THE SAME TICKER — the board is arguing with itself; net it before trading (ticker view →)
NVDA · Nvidia risk: special-purpose silicon + Huawei SHORT DORMANT 10.4

also touching these tickers, same direction: AI compute capex supercycle has 12-24 months of runway 87 (NVDA) · Nvidia + Hugging Face makes NVDA the open-source AI champion 58 (NVDA)

⚖ Why this verdict

fully deterministic — evaluate.py replays this from daily closes; nothing below is editable or hand-set

  1. Window: 2025-01-04 → 2026-01-04 — first mention + 12-month horizon, then the window locks.
  2. The call: ▲ LONG NVDA (primary play). NVDA rose 26.4% over the window → direction-adjusted R = +26.4% (the call made money).
  3. Benchmark: holding SPY over the same window returned +16.1%α = +26.4 − (+16.1) = +10.3% — what this call made or lost against just owning the index. This is the number the verdict uses.
  4. Rule fired:
    ▶ HIT — R ≥ +10% AND α ≥ +5 ✓ (R 26.4, α 10.3)
    · PARTIAL — R ≥ +5% OR α ≥ 0
    · MISS — everything else
  5. Credit: supporters of a HIT earn 1.0 each, opposers the inverse — this feeds the scoreboard weights. supported: Cyan Banister, Gavin Baker, Sacks, Thomas Laffont

Conviction timeline

bands: green ≥ 65 · watch ≥ 45 · ember ≥ 15

Plays vs SPY · % since first mention (2025-01-04)

Plays

expressionsymbolkindrelevancerationale
▲ LONG NVDAstock PRIMARY named directly when Jason asked 'so you're long NVIDIA and chip makers?' — 'Absolutely'

Mention log

Gavin Baker
Gavin Baker (regular guest ×1.0) tech w=1.46 · n=13 · E209 (2025-01-04) · annual_prediction · strength 2 ▶ 21:22 SUPPORT · horizon 12mo
“Inference compute is going to be the derivative winner... we're going to run out of GPUs, accelerators, compute in 2025 the same way we did in 2023. [Jason: 'So you're also long NVIDIA and chip makers?'] Absolutely.”
CY
Cyan Banister (regular guest ×1.0) tech: 4 resolved, unproven · E210 (2025-01-11) · explicit_prediction · strength 2 ▶ 1:21:19 SUPPORT
“We haven't even barely touched what AI is going to do and change and all of the various things that are going to come from it. And the early adopters cannot use Cloud without getting shut down because of scaling issues. And I don't think those are artificially created based on the type of investing I'm doing. And so I'm very bullish on Nvidia.”
Sacks
Sacks tech w=0.77 · n=53 · E213 (2025-01-31) · explicit_prediction · strength 2 ▶ 46:28 SUPPORT POLICY-ADJACENT
“Yeah, so there's a thing called Jevons Paradox, which kind of speaks to this concept. Satya actually tweeted about it, which is the economic concept where as the cost of a particular use goes down, the aggregate demand for all consumption of that thing goes up. So the basic idea is that as the price of AI gets cheaper and cheaper, we're going to want to use more and more of it. So you might actually get more spending on it in the aggregate.”
Gavin Baker
Gavin Baker (regular guest ×1.0) tech w=1.46 · n=13 · E221 (2025-03-29) · explicit_prediction · strength 3 ▶ 11:07 SUPPORT
“So we're definitely not overbuilt yet. Like, since Deepsea came out, like, what has happened since Deepsea Car 1 came out as China is buying every GPU they can? One of the largest Chinese server manufacturers warned last night that there's about to be a shortage of GPUs in China. The price of memory, DRAM, goes up every day.”
PH
Philippe Laffont (guest ×0.5) no tech track record · E227 (2025-05-09) · explicit_prediction · strength 3 ▶ 28:47 SUPPORT
“CapEx is going up and everybody has a gigantic shortage of chips right now. That I know for sure from all of our private companies, public companies, there's a shortage of chip, a shortage of compute power.”
TH
Thomas Laffont (regular guest ×1.0) tech w=1.17 · n=6 · E232 (2025-06-21) · explicit_prediction · strength 2 ▶ 29:58 SUPPORT
“Look, I think to me, number one, I still think NVIDIA, right? I don't see the GPU kind of getting displaced. I see additional architectures kind of coming on board, right? And growing the market. But at the end of the day, all roads still lead to the GPU for all of these models.”
Gavin Baker
Gavin Baker (regular guest ×1.0) tech w=1.46 · n=13 · E238 (2025-08-09) · explicit_prediction · strength 2 ▶ 19:28 SUPPORT
“So all of the dollars that are going into the ground are being used. We're not building ahead of demand. In fact, we're clearly still behind demand.”

Who built this conviction

each voice's total force on the score — supports and opposes from every mention, weighted exactly as the replay applied them · share = % of all mention-driven movement

Gavin Baker
Gavin Baker w=1.46 regular guest ×1.0
3 scoring events · 68% of moves
+78.5 → net +78.5
TH
Thomas Laffont w=1.17 regular guest ×1.0
1 scoring event · 11% of moves
+12.7 → net +12.7
CY
Cyan Banister regular guest ×1.0
1 scoring event · 9% of moves
+10.1 → net +10.1
Sacks
Sacks w=0.77
1 scoring event · 6% of moves
+7.0 → net +7.0
PH
Philippe Laffont guest ×0.5
1 scoring event · 6% of moves
+6.5 → net +6.5

⏳ decay drained -84.7 over the idea's life — that's time passing, attributed to no one

Score events

episodekindΔafternote
E209 2025-01-04 init +55.0 55.0 E209 born by Gavin Baker (annual_prediction x2) [w=1.46]
E210 2025-01-11 reinforce +10.1 65.1 E210 Cyan Banister support x2 (new voice)
E211 2025-01-18 decay -2.6 62.5 E211 silent
E212 2025-01-25 decay -2.5 60.0 E212 silent
E213 2025-01-31 reinforce +7.0 67.0 E213 Sacks support x2 (new voice) [w=0.77]
E214 2025-02-07 decay -2.7 64.3 E214 silent
E215 2025-02-15 decay -2.6 61.7 E215 silent
E216 2025-02-21 decay -2.5 59.3 E216 silent
E217 2025-03-01 decay -2.4 56.9 E217 silent
E218 2025-03-08 decay -2.3 54.6 E218 silent
E219 2025-03-15 decay -2.2 52.4 E219 silent
E220 2025-03-22 decay -2.1 50.3 E220 silent
E221 2025-03-29 reinforce +13.1 63.4 E221 Gavin Baker support x3 [w=1.46]
E222 2025-04-05 decay -2.5 60.9 E222 silent
E223 2025-04-11 decay -2.4 58.4 E223 silent
E224 2025-04-19 decay -2.3 56.1 E224 silent
E225 2025-04-26 decay -2.2 53.8 E225 silent
E226 2025-05-02 decay -2.2 51.7 E226 silent
E227 2025-05-09 reinforce +6.5 58.2 E227 Philippe Laffont support x3 (new voice) [w=0.50]
E228 2025-05-17 decay -2.3 55.9 E228 silent
E229 2025-05-24 decay -2.2 53.6 E229 silent
E231 2025-06-13 decay -2.1 51.5 E231 silent
E232 2025-06-21 reinforce +12.7 64.2 E232 Thomas Laffont support x2 (new voice) [w=1.17]
E233 2025-06-28 decay -2.6 61.7 E233 silent
E234 2025-07-04 decay -2.5 59.2 E234 silent
E235 2025-07-11 decay -2.4 56.8 E235 silent
E236 2025-07-19 decay -2.3 54.6 E236 silent
E237 2025-08-01 decay -2.2 52.4 E237 silent
E238 2025-08-09 reinforce +10.4 62.8 E238 Gavin Baker support x2 [w=1.46]
E239 2025-08-15 decay -2.5 60.3 E239 silent
E240 2025-08-22 decay -2.4 57.9 E240 silent
E241 2025-08-29 decay -2.3 55.6 E241 silent
E242 2025-09-07 decay -2.2 53.4 E242 silent
E243 2025-09-19 decay -2.1 51.2 E243 silent
E244 2025-09-27 decay -2.0 49.2 E244 silent
E245 2025-10-03 decay -2.0 47.2 E245 silent
E246 2025-10-10 decay -1.9 45.3 E246 silent
E247 2025-10-17 decay -1.8 43.5 E247 silent
E248 2025-10-24 decay -1.7 41.8 E248 silent
E249 2025-10-31 decay -1.7 40.1 E249 silent
E250 2025-11-07 decay -1.6 38.5 E250 silent
E251 2025-11-14 decay -1.5 36.9 E251 silent
E252 2025-11-22 decay -1.5 35.5 E252 silent
E253 2025-12-06 decay -1.4 34.1 E253 silent
E254 2025-12-13 decay -1.4 32.7 E254 silent
E255 2025-12-19 decay -1.3 31.4 E255 silent
E256 2025-12-31 decay -1.3 30.1 E256 silent