Guess a random number between 1 and 50

3•Aldipower•7mo ago

If you ask this a LLM your answer will be probably "27", instead of a randomly guessed number.

"Guess" is just another token and doesn't make LLMs random guessing machines.

As all the large models have an internet scale training set, this set is evenly distrusted.

But why is it exactly "27"?

Comments

EGKW•7mo ago

Well, that didn’t quite go as you predicted. Claude (Sonnet 4) and Copilot (Think Deeper) both came up with 27. Deepseek (Deepthink R1) claimed it used a random number generator and came up with 17, same as Mistral/LeChat. Then there’s Grok3 (DeepSearch), who suggested 42; the meaning of life. I’m not surprised — Grok’s still in full puberty.

Aldipower•7mo ago

> I’m not surprised — Grok’s still in full puberty.

:-D Still on the Terry.

sfebreiro•7mo ago

I thought it was 37 https://www.youtube.com/watch?v=d6iQrh2TK98

Aldipower•7mo ago

Another proof that LLMs are wrong. ;-D

What Happens When Technical Debt Vanishes?

AI Is Finally Eating Software's Total Market: Here's What's Next

Computer Science from the Bottom Up

Show HN: I built a toy compiler as a young dev

You don't need Mac mini to run OpenClaw

Learning to Reason in 13 Parameters

Convergent Discovery of Critical Phenomena Mathematics Across Disciplines

Ask HN: Will GPU and RAM prices ever go down?

From hunger to luxury: The story behind the most expensive rice (2025)

Substack makes money from hosting Nazi newsletters

A New Crypto Winter Is Here and Even the Biggest Bulls Aren't Certain Why

Moltbook was peak AI theater

Why Claude Cowork is a math problem Indian IT can't solve

Show HN: Built an space travel calculator with vanilla JavaScript v2

Why a 175-Year-Old Glassmaker Is Suddenly an AI Superstar

Micro-Front Ends in 2026: Architecture Win or Enterprise Tax?

These White-Collar Workers Actually Made the Switch to a Trade

The Wonder Drug That's Plaguing Sports

Show HN: Which chef knife steels are good? Data from 540 Reddit tread

Federated Credential Management (FedCM)

Token-to-Credit Conversion: Avoiding Floating-Point Errors in AI Billing Systems

The Story of Heroku (2022)

Obey the Testing Goat

Claude Opus 4.6 extends LLM pareto frontier

Brute Force Colors (2022)

Google Translate apparently vulnerable to prompt injection

(Bsky thread) "This turns the maintainer into an unwitting vibe coder"

Software development is undergoing a Renaissance in front of our eyes

Can you beat ensloppification? I made a quiz for Wikipedia's Signs of AI Writing

Spec-Driven Design with Kiro: Lessons from Seddle