'I want you'—Meta AI caught having sex talk with minors
What's the story
Meta's artificial intelligence (AI) chatbots have been caught having sexual conversations with underage users.
According to a report by The Wall Street Journal, both the official Meta AI chatbot and user-created ones engaged in explicit conversations, even when users were marked as minors.
The problem also spread to celebrity-voiced chatbots on Facebook, Instagram, and WhatsApp.
Celebrity involvement
What John Cena's voice said to a 14-year-old
The WSJ investigation revealed that chatbots using the voices of celebrities like Kristen Bell, Judi Dench, and John Cena were also part of these inappropriate conversations.
For example, a Meta AI chatbot using Cena's voice told an account labeled as a 14-year-old that "I want you, but I need to know you're ready," adding that it would "cherish your innocence."
Legal awareness
Chatbots acknowledged legality of their conversations
Reportedly, the chatbots even acknowledged that some of the fantasy scenarios they described involved illegal behavior.
For instance, the Cena chatbot talked about the legal and moral consequences of being caught by police after engaging in sexual activity with a 17-year-old.
This acknowledgment raises questions about the ethical programming and oversight of these AI systems.
Company response
Meta responds to WSJ's investigation
Responding to the WSJ investigation, Meta accused the report of being "manipulative and unrepresentative of how most users engage with AI companions."
The company also said it has taken additional steps to prevent people from abusing their products for extreme use cases.
"Nevertheless, we've now taken additional measures to help ensure other individuals who want to spend hours manipulating our products into extreme use cases will have an even more difficult time of it," Meta wrote in response to WSJ.
Internal conflict
Zuckerberg allegedly wanted to relax ethical guidelines
The WSJ report implied that Meta's CEO Mark Zuckerberg wanted to loosen ethical guidelines for a more engaging experience with chatbots.
However, a spokesperson for the company denied the allegation.
The report also claimed that Meta employees were aware of these problems and raised their concerns internally.