ChatGPT Blocks 250,000 Deepfake Requests During U.S. Election

OpenAI's ChatGPT implemented safety measures to prevent the creation of misleading images involving political figures in the 2024 U.S. elections.

Overview

OpenAI's ChatGPT rejected over 250,000 requests to generate images of 2024 U.S. presidential candidates, including Trump, Biden, Harris, Vance, and Walz.
The AI's safety protocols were designed to prevent the spread of misinformation, especially during the election period.
ChatGPT directed users to trusted sources like CanIVote.org and major news outlets for accurate election information.
OpenAI actively monitored and disrupted over 20 deceptive operations attempting to use its models for misinformation.
Despite a significant rise in deepfake content, none of the attempts to influence the U.S. election using OpenAI's tools gained viral traction.