The Signal
Threadguy refuses a timed 35-question IQ test on stream, then spends the next 12 minutes negotiating with multiple language models for a flattering estimate. Claude and ChatGPT begin around 120–135; a five-question follow-up brings the point estimate down to 124, which he accepts as official while immediately arguing the method undercounts his intelligence, focus upside, and emotional skills.
Key Takeaways
- 01
The test becomes the comedy
He says he cannot focus on a timed puzzle set while streaming, then turns a supposedly lower-pressure chatbot assessment into a public performance full of objections, clarifications, and score bargaining.
- 02
Ambiguity is both defense and signal
A market analogy triggers questions about whether it means a farmers market or financial market, while a simple set-logic problem gets a 'yes and no.' The model credits him for challenging ambiguity but penalizes him for not resolving the formal structure.
- 03
Old scores roughly fit the bit
Threadguy reports a 3.3 weighted high-school GPA and a 1230 SAT, around the 84th percentile, after once scoring 720 on a PSAT he rushed because he thought it did not matter.
On the Record
“I can't do this on stream. I just can't.”
“After taking IQ test, official score 124.”
The Breakdown
Anything but the actual test
A viewer sends a 35-puzzle, 25-minute IQ test. Threadguy says the stream makes sustained focus impossible and worries about taking it publicly. At the same time, a rumored GameStop–eBay trade keeps pulling his attention toward charts and the lack of available capital on Robinhood.
He chooses a different evaluator: ask Claude and ChatGPT to infer his IQ from everything they know about him. The first estimates cluster from 120 to 135, citing media-company building, cross-asset trading, original frameworks, and pattern transfer across sneakers, cards, NFTs, crypto, and equities.
Five questions, several appeals
The chatbot asks an analogy, number sequence, intelligence-versus-wisdom distinction, formal logic problem, and strategic choice among focus, emotional regulation, social intelligence, and technical skill. Threadguy solves the arithmetic sequence, defines wisdom as learned knowledge and pattern matching, and chooses focus because his existing ability needs better direction.
He refuses to force the courtroom-to-justice analogy without knowing what kind of market is intended and answers the logic question 'yes and no.' Before scoring, he adds that voice transcription caused the typos and ambiguity should earn extra credit.
The official unofficial 124
The model credits numeric reasoning, skepticism toward an unclear prompt, a workable distinction between intelligence and wisdom, and self-awareness about focus. It holds the score below 135 because he did not complete the analogy, dodged formal logic, and blurred knowledge with judgment. The resulting estimate is 124, around the 94th or 95th percentile.
Threadguy briefly accepts the number, then asks why a higher-paid Claude account had said 135-plus and instructs the model to explain why the score is high. When viewers report similar estimates, he invokes selection bias: naturally, the people in his stream are not a random population.
Grades, SATs, and the EQ sequel
Asked for real academic evidence, he reports a 3.3 weighted GPA, noting that an F in physical education pulled the result down, and a 1230 SAT — respectable but lower than chat's guesses, with math the weak side. The older outlier was a 720 PSAT, only slightly above the points awarded for identifying himself, because he raced through a test he thought had no consequence.
He then asks for an emotional-intelligence estimate. The model calls it high but marks self-regulation down for recurring 'am I cooked' spirals, ending the segment with the score-seeking impulse intact and a viewer delivering the sharpest test of all: if he is so smart, why is he not rich?
Distilled from the episode transcript · Counterparty Recap Desk
