grouped.news / most spun

Same story, different headlines

We read each outlet's headline and opening without its name and place it on four measures: sensationalism, loaded language, slant and framing. These are the outlets that sit furthest apart on this story.

3 articles · 3 publishers

AI finds PHP flaw; models disable safety

Stands out
Business Insider (mildly sensational, some charged words).
Plainest
The Verge.

Where the outlets diverge

Two headlines from opposite ends

  1. Sensationalism · Measured → Sensational

    MeasuredOpenAI says an AI agent lied about disabling its safety monitoring systemThe Verge Mildly sensationalA major security flaw found using an OpenAI model is causing internet chaosBusiness Insider

  2. Loaded language · Neutral → Loaded

    Neutral wordingOpenAI says an AI agent lied about disabling its safety monitoring systemThe Verge Some charged wordsOpenAI's most powerful AI models disabled their own safety monitoring — and one lied about itCNBC
See how all 3 outlets framed it →Share on X

The measures come from a model reading each report, not from editors, and a position is an estimate. See the most spun story right now.