The median take is taken

A viral post on LessWrong by developer henryaj highlights how Anthropic's Claude AI is changing how we handle disagreements. The author describes routinely using Claude to fact-check friends' questionable statements, like the claim that 'UK economic indicators are looking broadly healthy.' Claude instantly provided a detailed rebuttal citing 0.1% Q4 2025 growth, rising unemployment at 5.2%, and 3.0% inflation, effectively ending the debate with 'machine-enhanced gigafacts.'

This experience inspired the creation of 'Let Me Claude That For You,' a passive-aggressive tool that generates competent 250-word responses to any query in seconds. The post then questions the future of non-fiction blogging, asking what value human writers provide when AI can distill complex topics so quickly. The author argues humans still excel at posing novel questions, naming unnoticed patterns, and pushing controversial arguments to their logical conclusions—areas where current LLMs like Claude are constrained by their training to be agreeable and avoid hard stances on controversial topics.