← pool

🤖 Only foundation models can disrupt search - and Google is too far ahead

Chamath's read two weeks before ChatGPT shipped: the one thing that can displace Google search is a leap at the fundamental model level - powerful multimodal models - but Google is already so far ahead of everybody else on those models, and is the only player that can self-fund the billions the next leapfrog costs, while OpenAI will always have to raise from others. So the search franchise survives the AI transition, the technical moat is worth funding, and Larry Page's premise for diversifying Alphabet into Other Bets (that the core ad engine would be disrupted) was simply wrong.

0 CONVICTION
peaked 60.7 WATCH
EMBER
band · contested
▲ LONG GOOGL
expression · bullish tech
HIT
outcome · R +41.5% · α +25.5%
2023-11-19
window closed
⚠ CONFLICTING IDEAS ON THE SAME TICKER — the board is arguing with itself; net it before trading (ticker view →)
GOOGL · Blue cities ban or license-cap robotaxis SHORT EMBER 28.2
GOOGL · Tort floodgates open on social media SHORT DORMANT 14.6
GOOGL · ChatGPT takes Google search share and the market starts pricing the decay SHORT DORMANT 9.3

also touching these tickers, same direction: Google wins consumer AI via distribution 61 (GOOGL) · Anthropic IPO (~Oct 2026) is attractively priced at $1.5-2T 54 (GOOGL) · Personal AI agents: Google's data moat wins the form factor 29 (GOOGL) · Frontier AI labs keep durable premium pricing (not commoditized) 27 (GOOGL) · Own AWS, Azure and GCP and nothing else — the clouds capture the AI application dollars 21 (GOOGL) · Independent frontier-model labs lose to hyperscaler capital 14 (GOOGL) · Willow's error-correction scaling puts encryption on a two-to-five-year clock 9 (GOOGL) · Meta's AI assistant takes ten points of search share from Google 7 (GOOGL)

⚖ Why this verdict

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

  1. Window: 2022-11-19 → 2023-11-19 — first mention + 12-month horizon, then the window locks.
  2. The call: ▲ LONG GOOGL (primary play). GOOGL rose 41.5% over the window → direction-adjusted R = +41.5% (the call made money).
  3. Benchmark: holding SPY over the same window returned +16.0%α = +41.5 − (+16.0) = +25.5% — 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 41.5, α 25.5)
    · 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: Chamath, Friedberg, Jason | opposed: Sacks

Conviction timeline

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

Plays vs SPY · % since first mention (2022-11-19)

Plays

expressionsymbolkindrelevancerationale
▲ LONG GOOGLstock PRIMARY the franchise whose AI-disruption risk is the claim; Chamath argues the model lead and self-funding capacity protect it

Mention log

Chamath
Chamath tech w=0.90 · n=83 · E105 (2022-11-19) · explicit_prediction · strength 2 ▶ 1:11:47 SUPPORT
“Now, it probably stands to reason that if we have enough innovation at the fundamental model level in AI, particularly like a bunch of really powerful multimodal models, the new form of search can disrupt Google. But the problem is they are so ahead of everybody else with respect to those models as well.”
Jason
Jason tech w=1.05 · n=56 · E109 (2022-12-24) · explicit_prediction · strength 3 ▶ 28:28 OPPOSE
“Google, one of the greatest businesses and franchises ever created, has no answer currently for Chat GPT, because Google's business model is to get you to click on an ad between links.”
Chamath
Chamath tech w=0.90 · n=83 · E111 (2023-01-13) · explicit_prediction · strength 2 ▶ 35:07 SUPPORT
“I think that Google will open source their models because the most important thing that Google can do is reinforce the value of search. And the best way to do that is to scorch the earth with these models, which is to make them widely available and as free as possible.”
Friedberg
Friedberg tech w=1.08 · n=65 · E113 (2023-01-27) · sentiment · strength 2 ▶ 34:11 SUPPORT
“Show me where someone's made a better competitor to Google search. Show me where consumers don't choose to go to another search engine because Google's built a better search engine.”
Sacks
Sacks tech w=0.77 · n=53 · E113 (2023-01-27) · sentiment · strength 1 ▶ 34:19 SUPPORT
“No, it's because there's a data network effect there, that the more searches they provide, the more data they get. It's the reinforcement learning.”
Chamath
Chamath tech w=0.90 · n=83 · E115 (2023-02-11) · explicit_prediction · strength 3 ▶ 1:23:55 REVERSAL
“Google should go and they should cannibalize their own business before it is cannibalized for them.”
Jason
Jason tech w=1.05 · n=56 · E122 (2023-03-31) · explicit_prediction · strength 3 ▶ 32:10 SUPPORT
“And then when they make BARD a default, you know, a little snippet on your Google search return page or BARD is built into YouTube or Chrome or Android or the Play Store, they're going to roll right over ChatGPT because they have billions of users already.”
Chamath
Chamath tech w=0.90 · n=83 · E122 (2023-03-31) · sentiment · strength 2 ▶ 33:14 SUPPORT
“Google's incentive is to usurp ChatGPT's usage by inserting something inside of their existing distribution channels to suppress the ability for you to want to go to the app.”
Friedberg
Friedberg tech w=1.08 · n=65 · E126 (2023-04-28) · explicit_prediction · strength 2 ▶ 3:23 SUPPORT
“But Google certainly has an incredible set of tools and capabilities that is leap years ahead of Microsoft. They're in a position to really compete. They just have to have the will and the leadership to do it.”
Friedberg
Friedberg tech w=1.08 · n=65 · E128 (2023-05-12) · explicit_prediction · strength 3 ▶ 16:43 SUPPORT
“So if you can organize those people and they've got this significantly advantaged infrastructure, they have a real shot at being a platform player here.”
Sacks
Sacks tech w=0.77 · n=53 · E128 (2023-05-12) · sentiment · strength 3 ▶ 13:38 OPPOSE
“But I'm not finding the answers to be more accurate, and I'm not finding them to be more detailed. I'm not seeing a reason to use this over ChatGBT.”
Chamath
Chamath tech w=0.90 · n=83 · E128 (2023-05-12) · explicit_prediction · strength 2 ▶ 14:54 SUPPORT
“And so if they take Bard and they have the confidence now to just integrate it into Gmail, or integrate it into these other points where they already have hundreds of millions of users, that's like a really tough distribution barrier to overcome.”
Jason
Jason tech w=1.05 · n=56 · E128 (2023-05-12) · sentiment · strength 1 ▶ 14:50 SUPPORT
“I thought they were going to catch up real quick, and this seems like we got a race on our hands now.”
Jason
Jason tech w=1.05 · n=56 · E133 (2023-06-16) · explicit_prediction · strength 3 ▶ 47:45 SUPPORT · horizon 6mo
“I think Google is going to be, ChatGPT4, I'm saying it right here right now, I think they're going to beat them because I think that they're better at indexing all this information and understanding it than anybody on the planet. And they have the largest ad network. If they get this done in the next six months, I think it's going to increase the cost per click because they're going to know so much about each user”
Jason
Jason tech w=1.05 · n=56 · E147 (2023-09-29) · sentiment · strength 2 ▶ 1:12:42 SUPPORT
“So this idea that Google is going to be displaced or they're moving slow, that might be antiquated information.”

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
5 scoring events · 44% of moves
+60.4 / -28.8 → net +31.7
Jason
Jason w=1.05
5 scoring events · 28% of moves
+43.1 / -15.2 → net +27.9
Friedberg
Friedberg w=1.08
3 scoring events · 19% of moves
+38.4 → net +38.4
Sacks
Sacks w=0.77
2 scoring events · 9% of moves
+8.1 / -11.1 → net -3.0

⏳ decay drained -55.4 over the idea's life — that's time passing, attributed to no one

Score events

episodekindΔafternote
E105 2022-11-19 init +31.6 31.6 E105 born by Chamath (explicit_prediction x2) [w=0.90]
E106 2022-12-03 decay -1.3 30.4 E106 silent
E107 2022-12-10 decay -1.2 29.1 E107 silent
E108 2022-12-16 decay -1.2 28.0 E108 silent
E109 2022-12-24 oppose -15.2 12.8 E109 Jason opposes x3 [w=1.05]
E110 2023-01-06 decay -0.5 12.3 E110 silent
E111 2023-01-13 reinforce +11.9 24.2 E111 Chamath support x2 [w=0.90]
E112 2023-01-20 decay -1.0 23.2 E112 silent
E113 2023-01-27 reinforce +18.6 41.8 E113 Friedberg support x2 (new voice) [w=1.08]
E113 2023-01-27 reinforce +8.1 49.9 E113 Sacks support x1 (new voice) [w=0.77]
E114 2023-02-04 decay -2.0 47.9 E114 silent
E115 2023-02-11 reversal -28.8 19.2 E115 Chamath flips
E116 2023-02-17 decay -0.8 18.4 E116 silent
E118 2023-03-03 decay -0.7 17.7 E118 silent
E119 2023-03-11 decay -0.7 17.0 E119 silent
E120 2023-03-17 decay -0.7 16.3 E120 silent
E121 2023-03-24 decay -0.7 15.6 E121 silent
E122 2023-03-31 reinforce +16.0 31.6 E122 Jason support x3 (flipped from oppose) [w=1.05]
E122 2023-03-31 reinforce +9.3 40.9 E122 Chamath support x2 (flipped from reversal) [w=0.90]
E123 2023-04-07 decay -1.6 39.3 E123 silent
E124 2023-04-14 decay -1.6 37.7 E124 silent
E125 2023-04-21 decay -1.5 36.2 E125 silent
E126 2023-04-28 reinforce +10.3 46.5 E126 Friedberg support x2 [w=1.08]
E128 2023-05-12 oppose -11.1 35.4 E128 Sacks opposes x3 [w=0.77]
E128 2023-05-12 reinforce +8.2 43.5 E128 Jason support x1 [w=1.05]
E128 2023-05-12 reinforce +7.7 51.2 E128 Chamath support x2 [w=0.90]
E128 2023-05-12 reinforce +9.5 60.6 E128 Friedberg support x3 [w=1.08]
E129 2023-05-19 decay -2.4 58.2 E129 silent
E130 2023-05-26 decay -2.3 55.9 E130 silent
E131 2023-06-02 decay -2.2 53.7 E131 silent
E132 2023-06-10 decay -2.1 51.5 E132 silent
E133 2023-06-16 reinforce +9.2 60.7 E133 Jason support x3 [w=1.05]
E134 2023-06-24 decay -2.4 58.3 E134 silent
E135 2023-07-01 decay -2.3 55.9 E135 silent
E136 2023-07-09 decay -2.2 53.7 E136 silent
E137 2023-07-14 decay -2.1 51.6 E137 silent
E139 2023-07-27 decay -2.1 49.5 E139 silent
E140 2023-08-04 decay -2.0 47.5 E140 silent
E141 2023-08-11 decay -1.9 45.6 E141 silent
E142 2023-08-18 decay -1.8 43.8 E142 silent
E143 2023-08-25 decay -1.8 42.0 E143 silent
E144 2023-09-01 decay -1.7 40.4 E144 silent
E146 2023-09-22 decay -1.6 38.7 E146 silent
E147 2023-09-29 reinforce +9.7 48.4 E147 Jason support x2 [w=1.05]
E148 2023-10-07 decay -1.9 46.5 E148 silent
E149 2023-10-13 decay -1.9 44.6 E149 silent
E150 2023-10-20 decay -1.8 42.8 E150 silent
E151 2023-10-27 decay -1.7 41.1 E151 silent
E152 2023-11-03 decay -1.6 39.5 E152 silent