What happens during pretraining for an AI like Claude?
AHumans rate which replies are helpful or harmful
BModel reads vast text and learns language patterns
CEngineers write rules for how the model must reply
DThe model is tested against a fixed question bank
Answer & Solution
Correct answer: B. Model reads vast text and learns language patterns
Card 2 of this block: pretraining means reading billions of sentences from the internet, books, and web text so the model learns statistical language patterns. Source: simple_wikipedia_ai_ocr.md (machine learning section) + claude_constitution_ocr.md (role of human text in Claude's training, p.1).
Related questions
A friend asks: what should you do first when an AI answer contains a specific date?Which type of claim carries the highest hallucination risk?What is the main reason AI sometimes produces confident but wrong answers?You told AI your name yesterday. In a new chat today, AI does not know it. Why?AI says an essay has 850 words. The real count is 1003. What does this show?You ask AI for today's cricket score. Why can't it answer reliably?What does AI do well when you ask it to compare two options?Which task is AI best suited for?