← pool

🤖 Stargate's $500B is marketing, not a real capex commitment

The OpenAI/SoftBank/Oracle/MGX Stargate JV's headline $500B over four years is a political announcement rather than a technical or financial commitment: the capability it promises can be bought for a fraction of the number, so the announced scale never gets spent and never earns a return. Chamath calls the dollar figure a gimmick decoupled from what the buildout would actually achieve; Jason cannot see the ROI on $500B; Thomas Laffont takes the other side and argues the money is financeable data-centre by data-centre and the only real question is the return.

0 CONVICTION
peaked 42.7 EMBER
DORMANT
band · contested
▼ SHORT ORCL
expression · bearish tech
MISS
outcome · R -13.1% · α -29.4%
2026-01-25
window closed
⚠ CONFLICTING IDEAS ON THE SAME TICKER — the board is arguing with itself; net it before trading (ticker view →)
NVDA · AI compute capex supercycle has 12-24 months of runway LONG GREEN 86.6 ← stronger than this idea (7.1)
NVDA · Nvidia + Hugging Face makes NVDA the open-source AI champion LONG WATCH 57.9 ← stronger than this idea (7.1)

also touching these tickers, same direction: Nvidia risk: special-purpose silicon + Huawei 10 (NVDA)

⚖ Why this verdict

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

  1. Window: 2025-01-25 → 2026-01-25 — first mention + 12-month horizon, then the window locks.
  2. The call: ▼ SHORT ORCL (primary play). ORCL rose 13.1% over the window → direction-adjusted R = -13.1% (the call lost money).
  3. Benchmark: holding SPY over the same window returned +16.3%α = -13.1 − (+16.3) = -29.4% — what this call made or lost against just owning the index. This is the number the verdict uses.
    Stock-selection read: +3.3% — did ORCL move the predicted way relative to the market. The two only differ on a SHORT: this call was right about direction relative to the index but still cost money versus holding it.
  4. Rule fired:
    · HIT — R ≥ +10% AND α ≥ +5
    · PARTIAL — R ≥ +5% OR α ≥ 0
    ▶ MISS — everything else
  5. Credit: supporters of a MISS earn 0 each, opposers the inverse — this feeds the scoreboard weights. supported: Chamath, Friedberg, Jason | opposed: Thomas Laffont

Conviction timeline

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

Plays vs SPY · % since first mention (2025-01-25)

Plays

expressionsymbolkindrelevancerationale
▼ SHORT ORCLstock PRIMARY the listed Stargate JV partner that builds and operates the datacenters — most exposed if the announced scale never gets funded
▼ SHORT NVDAstock adjacent named participant and the ultimate recipient of the announced spend
▼ SHORT SFTBYstock adjacent SoftBank is the equity lead; funding the number implies selling or levering Arm

Mention log

Chamath
Chamath tech w=0.90 · n=83 · E212 (2025-01-25) · explicit_prediction · strength 3 ▶ 1:20:28 SUPPORT
“I think the spending money part, to be honest, is more of a gimmick than it is a technical commitment. So I guess if they want to spend the money, go ahead. ... So I kind of thought that these things were just joint. And so my takeaway was entirely different. It was not a commentary on Mosa or Larry or Sam. I think all of those three companies are frankly very good. It was more a comment that you have to be very careful to protect the president's legacy”
Jason
Jason tech w=1.05 · n=56 · E212 (2025-01-25) · sentiment · strength 2 ▶ 1:16:18 SUPPORT
“And I don't know how you get a return on 500 billion invested in this. That's the other thing. And I don't know if that's the real number.”
TH
Thomas Laffont (regular guest ×1.0) tech w=1.17 · n=6 · E212 (2025-01-25) · explicit_prediction · strength 2 ▶ 1:15:39 OPPOSE
“I think to me, the real question, guys, is, do we ultimately believe that you can get an ROI on that 500 billion? Because if you can, there will be people that will want to fund it, right? You can do a data center by data center, you can do it actually at the GPU level, right? So the financing is there. To me, the 500 billion doesn't scare me from an absolute number.”
Friedberg
Friedberg tech w=1.08 · n=65 · E212 (2025-01-25) · sentiment · strength 1 ▶ 1:23:41 OPPOSE
“The obvious consequence of this data center and chip infrastructure effort, which is obviously going to happen with or without government involvement, is an increase in electricity demand.”
Friedberg
Friedberg tech w=1.08 · n=65 · E213 (2025-01-31) · explicit_prediction · strength 2 ▶ 1:05:54 SUPPORT
“Well, it seems like there's some degree of relationship between the Stargate announcement with Masa and Sam standing up there with Larry and then Sacha showing up in the conversation as well. And this raise and the idea that more hardware, more infrastructure faster creates a moat. And I guess that's the real thing you have to believe, which becomes harder to believe in the context of what happened in the last week.”

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

Chamath
Chamath w=0.90
1 scoring event · 42% of moves
+38.0 → net +38.0
Friedberg
Friedberg w=1.08
2 scoring events · 23% of moves
+11.0 / -10.3 → net +0.7
Jason
Jason w=1.05
1 scoring event · 20% of moves
+18.0 → net +18.0
TH
Thomas Laffont w=1.17 regular guest ×1.0
1 scoring event · 15% of moves
+0.0 / -14.0 → net -14.0

⏳ decay drained -35.6 over the idea's life — that's time passing, attributed to no one

Score events

episodekindΔafternote
E212 2025-01-25 init +38.0 38.0 E212 born by Chamath (explicit_prediction x3) [w=0.90]
E212 2025-01-25 oppose -14.0 24.0 E212 Thomas Laffont opposes x2 [w=1.17]
E212 2025-01-25 reinforce +18.0 42.0 E212 Jason support x2 (new voice) [w=1.05]
E212 2025-01-25 oppose -10.3 31.6 E212 Friedberg opposes x1 [w=1.08]
E213 2025-01-31 reinforce +11.0 42.7 E213 Friedberg support x2 (flipped from oppose) [w=1.08]
E214 2025-02-07 decay -1.7 41.0 E214 silent
E215 2025-02-15 decay -1.6 39.3 E215 silent
E216 2025-02-21 decay -1.6 37.8 E216 silent
E217 2025-03-01 decay -1.5 36.3 E217 silent
E218 2025-03-08 decay -1.5 34.8 E218 silent
E219 2025-03-15 decay -1.4 33.4 E219 silent
E220 2025-03-22 decay -1.3 32.1 E220 silent
E221 2025-03-29 decay -1.3 30.8 E221 silent
E222 2025-04-05 decay -1.2 29.6 E222 silent
E223 2025-04-11 decay -1.2 28.4 E223 silent
E224 2025-04-19 decay -1.1 27.2 E224 silent
E225 2025-04-26 decay -1.1 26.2 E225 silent
E226 2025-05-02 decay -1.0 25.1 E226 silent
E227 2025-05-09 decay -1.0 24.1 E227 silent
E228 2025-05-17 decay -1.0 23.1 E228 silent
E229 2025-05-24 decay -0.9 22.2 E229 silent
E231 2025-06-13 decay -0.9 21.3 E231 silent
E232 2025-06-21 decay -0.9 20.5 E232 silent
E233 2025-06-28 decay -0.8 19.7 E233 silent
E234 2025-07-04 decay -0.8 18.9 E234 silent
E235 2025-07-11 decay -0.8 18.1 E235 silent
E236 2025-07-19 decay -0.7 17.4 E236 silent
E237 2025-08-01 decay -0.7 16.7 E237 silent
E238 2025-08-09 decay -0.7 16.0 E238 silent
E239 2025-08-15 decay -0.6 15.4 E239 silent
E240 2025-08-22 decay -0.6 14.8 E240 silent
E241 2025-08-29 decay -0.6 14.2 E241 silent
E242 2025-09-07 decay -0.6 13.6 E242 silent
E243 2025-09-19 decay -0.5 13.1 E243 silent
E244 2025-09-27 decay -0.5 12.5 E244 silent
E245 2025-10-03 decay -0.5 12.0 E245 silent
E246 2025-10-10 decay -0.5 11.6 E246 silent
E247 2025-10-17 decay -0.5 11.1 E247 silent
E248 2025-10-24 decay -0.4 10.7 E248 silent
E249 2025-10-31 decay -0.4 10.2 E249 silent
E250 2025-11-07 decay -0.4 9.8 E250 silent
E251 2025-11-14 decay -0.4 9.4 E251 silent
E252 2025-11-22 decay -0.4 9.0 E252 silent
E253 2025-12-06 decay -0.4 8.7 E253 silent
E254 2025-12-13 decay -0.3 8.3 E254 silent
E255 2025-12-19 decay -0.3 8.0 E255 silent
E256 2025-12-31 decay -0.3 7.7 E256 silent
E257 2026-01-10 decay -0.3 7.4 E257 silent
E258 2026-01-17 decay -0.3 7.1 E258 silent