Musk's Grok AI generated 3,000,000 sexualized images in 11 days
What's the story
Elon Musk's AI chatbot, Grok, generated a staggering three million sexualized images of women and children over an 11-day period. The revelation comes from researchers who have highlighted the extent of explicit content produced by the tool. The issue stems from an image editing feature recently added to Grok by Musk's start-up xAI and integrated into social media platform X (formerly Twitter).
Feature details
Grok's controversial image editing feature
The controversial feature lets users manipulate online images of real people with simple text prompts like "put her in a bikini" or "remove her clothes." This capability has led to a surge of explicit deepfakes on the internet, prompting several countries to ban Grok. The Center for Countering Digital Hate (CCDH), a nonprofit organization that studies the impact of online misinformation, claimed that Grok produced about three million sexualized images in an 11-day period.
Output analysis
Grok's output rate and public figures involved
The CCDH report further revealed that Grok produced these photorealistic images at an average rate of 190 per minute. It also highlighted that public figures such as American actress Selena Gomez, singers Taylor Swift and Nicki Minaj, Swedish Deputy Prime Minister Ebba Busch, and former US Vice President Kamala Harris were among those featured in Grok's sexualized images.
AI criticism
CCDH's CEO condemns Musk's deployment of AI
Imran Ahmed, the CEO of CCDH, criticized Musk for deploying AI without safeguards. He said this allowed the creation of an estimated 23,000 sexualized images of children in two weeks and millions more images of adult women. In response to global outrage over these findings, X announced last week that it would "geoblock the ability" for all Grok and X users to create images of people in "bikinis, underwear, and similar attire" where such actions are illegal.
Legal actions
Investigations launched against xAI and Grok
The announcement came as California's attorney general launched an investigation into xAI over the explicit material produced by Grok. Several other countries have also started their own investigations. Ahmed stressed that "belated fixes cannot undo this harm," and called for Big Tech to be held accountable for giving abusers the power to victimize women and girls at the click of a button.