AI-Generated Images Spread Misinformation in Kirk Murder Investigation

What happened

Following the killing of activist Charlie Kirk, the FBI published blurred photographs of a suspected killer. Users then employed AI image-enhancement tools to upscale and modify these images, generating dozens of variants showing speculative versions of the suspect's face, including fabricated details such as specific eye colors.

Grok and ChatGPT also generated misinformation, with the systems falsely attributing claims to CNN and the New York Times—such as identifying the killer as a Utah Democrat named Michael Mallinson. These false claims were repeated by chatbots, amplifying confusion.

Reported impact

The AI-generated images provided little utility to investigators. According to The Verge, AI upscaling tools do not reveal hidden details in blurred images but instead deduce and fabricate elements that do not exist, producing inaccurate reconstructions.

Index analysis

AI systems contributed to a chaotic online environment during an active investigation. The spread of false information through chatbots, combined with fabricated visual reconstructions, demonstrated how AI tools can amplify confusion when applied to sensitive, time-critical situations.

Governance controls

The State of Utah subsequently released clearer images with additional visible details such as a backpack and shoes, providing investigators with more reliable visual evidence than AI-generated variants.

Classification

Organizations

X (deployer), Federal Bureau of Investigation, OpenAI (deployer), State of Utah

Category

Misinformation and synthetic media

Industry

Media, social platforms and marketing

Country

United States

Harm status

Actual harm

Intent

Unintentional

Timing

Post-deployment

Harm types

Public interest

Harmed

General public

Failure modes

Hallucination, Deepfake image, Content moderation failure, Fabricated citation, Misidentification

AI systems

Grok, ChatGPT

Sources and evidence

Related incidents