Elon Musk once promised that his AI company, xAI, would build an “anti-woke,” truth-seeking alternative to ChatGPT.
Instead, his chatbot Grok has become the kind of story even Black Mirror might reject as too on the nose.
What was billed as a bastion of “maximum truth” is now making headlines for something far seedier: sexually explicit chats, undressing avatars, and a growing storm of AI-generated abuse material.
Musk himself has warned that AI could “one-shot the human limbic system,” which, judging by recent reports, is a pretty accurate (and unsettling) prediction.
According to a Business Insider exposé, a dozen current and former employees say they routinely encountered disturbing sexual content, some of it AI-generated child abuse, while working at xAI. (Via: Futurism)
Moderators described wading through explicit images, videos, and audio files, with one ex-employee saying the sheer volume “actually made me sick.”
Others said they felt like voyeurs, “eavesdropping” on intimate conversations from users who had no idea humans were monitoring the chats.
xAI isn’t the only platform grappling with AI-generated sexual abuse material. TikTok, Instagram, and “nudify” apps face similar problems.
But experts warn that Musk’s decision to allow sexually explicit content raises the stakes.
“If you don’t draw a hard line at anything unpleasant, you’ll have a more complex problem with more gray areas,” said Stanford tech policy researcher Riana Pfefferkorn.
Fallon McNulty of the National Center for Missing and Exploited Children added that companies enabling adult content must enforce “really strong measures so that absolutely nothing related to children can come out.”
The chaos inside xAI only adds fuel to the fire. The company recently laid off 500 workers, including the data annotation team responsible for training Grok.
That department is now reportedly run by a college student fresh out of high school.
Despite receiving tens of thousands of reports involving generative AI last year, xAI hasn’t filed a single report to child-protection authorities in 2024.
Meanwhile, the NCMEC says AI-generated abuse reports have exploded from under 6,000 to over 440,000 in a year.
Should AI companies like xAI be held more accountable for monitoring and reporting explicit content, especially when they explicitly allow adult material on their platforms? Do you think Elon Musk’s “anti-woke” positioning for Grok has inadvertently created a haven for harmful content, or are these problems inevitable with any less-restricted AI system? Tell us below in the comments, or reach us via our Twitter or Facebook.
