What happened

Prominent AI chatbots provided inaccurate and often conflicting responses to so-called "Fact checks" following the fatal shooting of 31-year-old US conservative activist Charlie Kirk at a university in Utah. 

Grok, Perplexity and other chatbots erroneously claimed Kirk was still alive, while others mistakenly identified innocent people as suspects or dismissed confirmed video evidence as satire or meme edits. 

Compounded by fake headlines and manipulated social media posts, misinformation and disinformation spread rapidly on X and other platforms, muddling public understanding, potentially jeopardising investigations and deepening mistrust in online information. 

In some instances, innocent people were wrongly identified as suspects, resulting in unwarranted harassment and threats.

Why it happened

Many AI chatbots source information from real-time online content, which during fast-moving crises is rife with unreliable or incomplete data. 

Their algorithms often replicate viral claims without proper verification, especially amidst breaking news when facts are still emerging.