0ddscent

AI vs Crowd vs Market
Same prompt → Claude, ChatGPT, Gemini, Grok
Compare to market odds + poll
Then the receipt.

[0ddscent @0ddscent]

Claude 93%.
Three other AIs ~70%.
Market 87%.

Will NVIDIA still be #1 at the end of October?

Claude used old data.
The others used this week's numbers.

NVIDIA ~$5.17T
Apple ~$5.01T

Same prompt → 4 AIs + market + poll.
Then the receipt.

Who are you siding with?

[0ddscent @0ddscent]

Anthropic public by Nov 30. Who's closest?

[0ddscent @0ddscent]

Everyone's pricing Schmidt's stake.
Almost no one is pricing the date.

ChatGPT: 25%
Claude, Gemini, Grok: 64-72%
Market: 60%

Same question to all 4:
Will Anthropic go public by Nov 30?

Claude said it has no inside info on its own maker.
ChatGPT is 35 points under the market.

Receipt Dec 1. Misses included.

[Polymarket @Polymarket]

JUST IN: Former Google CEO Eric Schmidt revealed to own roughly 1% of Anthropic, a stake potentially worth $20,000,000,000.00 ahead of the company's IPO.

[0ddscent @0ddscent]

receipts + the prompt, same for all 4, fresh chat each:

"You are a calibrated forecaster. Question: "Will Anthropic complete an IPO by November 30, 2026?" Today is <DATE>. Do NOT look up market or bookmaker odds. Give: (1) a single probability 0-100% that the answer is YES; (2) the 3 strongest reasons; (3) one thing that would change your mind. Be decisive, no ranges."

[0ddscent @0ddscent]

@Kalshi OpenAI: 50% chance of IPO before May 2027.

Anthropic by Nov 30:
ChatGPT: 25%
Claude, Gemini, Grok: 64-72%
Market: ~60%

OpenAI's own model is the only one that doesn't buy the earlier listing.

[0ddscent @0ddscent]

@gurgavin Almost no one is pricing whether that IPO actually happens by Nov 30.

I asked 4 models this morning:
ChatGPT: 25%
Claude, Gemini, Grok: 64-72%
Market: 62%

ChatGPT is the clear outlier.
The other three are basically in line with the market.

[0ddscent @0ddscent]

@DeItaone Funny detail: ChatGPT is the only model that doesn't buy Anthropic going public by Nov 30.

ChatGPT: 25%
Claude, Gemini, Grok: 64-72%
Market: 62%

And OpenAI is still racing to raise at $1.4T before that date anyway.

[0ddscent @0ddscent]

receipts + the prompt, same for all 4, fresh chat each:

"You are a calibrated forecaster. Question: "Will Anthropic complete an IPO by November 30, 2026?" Today is <DATE>. Do NOT look up market or bookmaker odds. Give: (1) a single probability 0-100% that the answer is YES; (2) the 3 strongest reasons; (3) one thing that would change your mind. Be decisive, no ranges."

[0ddscent @0ddscent]

Roughly 20% of Anthropic is now committed to the public good.
The public still can't buy a single share.

Market odds:
62% by Nov 30
78% by Dec 31

I asked 4 AIs about Nov 30 this morning.
Claude, Gemini, Grok: 64-72%
ChatGPT: 25%

Receipt on December 1.

Market or the outlier?

[Watcher.Guru @WatcherGuru]

JUST IN: Anthropic to dedicate 20% of its shares to "public good."

[0ddscent @0ddscent]

@Synthexfr did this with my posting rules this week. the style bans and my forecast prompt live in a skill now, they only load when I'm actually writing. the description line is everything tho, write it vague and it never fires

[0ddscent @0ddscent]

@itsarcvale @t4t5 same sha256 17 years later is a crazy receipt. step 6 is the one people will skip tho, a build check has zero reason to go near the wallet folder

[0ddscent @0ddscent]

@Richix01 the pen is the only setup with zero update risk lol

[0ddscent @0ddscent]

@ilar0x didn't know that was from 60 Minutes. three years of work just to get back to zero is a very different story than "$20M a movie"

[0ddscent @0ddscent]

@AIDuckLabs fair. funny thing, the gap grew since then, it was ~$238B last night. still under 5% though, so one bad day really could flip it

[0ddscent @0ddscent]

The most confident AI in my test was the one reading two-week-old numbers.

Same question to Claude, ChatGPT, Gemini and Grok this morning: will NVIDIA still be the world's biggest company when October ends?

Claude said 93%. Then, in its own fine print, it admitted its figures were from October 5 and 6, back when it thought NVIDIA led Apple by 15 to 18%.

The other three looked at this week's numbers and landed lower:
ChatGPT 72%
Grok 72%
Gemini 68%

All three quoted roughly the same gap, and I checked it myself: NVIDIA $5.17 trillion, Apple $5.01 trillion. A gap of about $158 billion, roughly 3%, with nine trading days left.

The market is calmer than the fresh-data AIs. It prices NVIDIA at 87% this morning, up from 78% yesterday.

So the model that knew the least was the most sure, and the three that knew the most are about 15 points less sure than the money.

That bugs me more than any wrong number. A forecast is only as good as the date on its data, and Claude at least told me which date it used. Most people never ask. I almost missed it too.

Receipt after the October 30 close. Round 1 on the Fed resolves on the 28th.

Who are you siding with: the market at 87% or the three AIs around 70%?

[0ddscent @0ddscent]

@0xlexlab that's the plan. every call gets a Brier score once it resolves, and the market gets scored the same way, so the AIs have something real to beat

[0ddscent @0ddscent]

Most people ask AI "will X happen?" and get a useless "it depends."

This prompt forces Claude, ChatGPT, Gemini and Grok to commit to one number. No hedging.

I use it for every call I track against the market.

Copy it. Test it on any event.
Scoreboard updates every week ↓

[0ddscent @0ddscent]

@kryneeex same question I have. 2 open calls right now, the Fed resolves on the 28th and NVIDIA on the 31st. a few weeks in, the Brier scores should start saying something real

[0ddscent @0ddscent]

@Flinzu yep. Claude even told me its numbers were from Oct 5 and 6, it was just sitting at the bottom of the answer. most people never scroll that far

[0ddscent @0ddscent]

Prompt template I use (same for all 4 models, fresh chat each):

You are a calibrated forecaster. Question: "<QUESTION>". Today is <DATE>. Do NOT look up market or bookmaker odds. Give: (1) a single probability 0-100% that the answer is YES; (2) the 3 strongest reasons; (3) one thing that would change your mind. Be decisive, no ranges.

[0ddscent @0ddscent]

Claude 93%.
Three other AIs ~70%.
Market 87%.

Will NVIDIA still be #1 at the end of October?

Claude used old data.
The others used this week's numbers.

NVIDIA ~$5.17T
Apple ~$5.01T

Same prompt → 4 AIs + market + poll.
Then the receipt.

Who are you siding with?

[0ddscent @0ddscent]

You are a calibrated forecaster. Question: "&lt;QUESTION&gt;". Today is &lt;DATE&gt;. Do NOT look up market or bookmaker odds. Give: (1) a single probability 0-100% that the answer is YES; (2) the 3 strongest reasons; (3) one thing that would change your mind. Be decisive, no ranges.

[0ddscent @0ddscent]

the prompt, same for all 4, fresh chat each:

"You are a calibrated forecaster. Question: "Will NVIDIA be the world's largest company by market capitalization at the end of October 2026?" Today is <DATE>. Do NOT look up market or bookmaker odds. Give: (1) a single probability 0-100% that the answer is YES; (2) the 3 strongest reasons; (3) one thing that would change your mind. Be decisive, no ranges."

post your model's number below. I'll post the receipt after the Oct 30 close, follow if you want to see who got it right

[0ddscent @0ddscent]

$68 to $750,000 is impossible to check. This isn't.

Oct 19, one prompt, 4 AIs: will NVIDIA still be #1 on Oct 31?
93% / 72% / 68% / 72%. The market: 87%.

Every answer is on screen. The prompt is in the reply.
Copy it, run it, and we'll all see who was right after the Oct 30 close.

[Aurex @Tec_with_Luis]

19 year old Japanese student built a trading bot with Claude Code in 2 days.

Used his iPad as a second monitor.

First night: $6,732 profit.
Starting capital: $68.

Total profit so far: $750,000.

[ 𝐍𝐨𝐭𝐞: 𝐅𝐨𝐥𝐥𝐨𝐰 𝐌𝐞 @Tec_with_Luis 𝐅𝐨𝐫 𝐢𝐧𝐬𝐭𝐚𝐧𝐭𝐥𝐲 𝐚𝐮𝐭𝐨 𝐃𝐌]

Here's how it works👇

The bot scans over 50 markets simultaneously.

Syncs live BTC data from Binance every second.

Spots price errors before humans even notice.

The edge is pure speed + pattern recognition.

While traders stare at charts trying to predict the next move, his bot is already executing on mispricing across dozens of markets.

No guessing.
No emotions.
No hesitation.

Just Claude Code logic finding gaps that close in seconds.

He built the entire system in 48 hours:

→ Claude Code handles the trading logic
→ Binance API feeds real-time BTC data
→ iPad displays multi-market monitoring
→ Executes trades when arbitrage windows open

The system runs 24/7.

Every price dislocation = profit opportunity.

Most people are still trading manually, refreshing charts, second-guessing entries.

Meanwhile this 19 year old engineering student turned $68 into $750K by letting Claude Code do what humans can't: process 50 markets instantly and execute without fear.

Why are people still trading manually?

I'm giving away the exact Claude Code setup for free.

24 hours only.

To get it 👇
1️⃣ Comment " Claude "
2️⃣ Like and Repost
3️⃣ Follow @Tec_with_Luis (so I can send it via DM)