NYC Council Speaker Accuses SpaceX's Grok of Generating 'Sexualized' Images
Julie Menin demonstrated how the AI chatbot produced inappropriate images of her using official headshots, despite company claims of safeguards.
New York City Council Speaker Julie Menin confronted a representative from Elon Musk's SpaceX on Friday, accusing the company's AI chatbot, Grok, of generating "sexualized" images of users despite assurances of safety measures.
Menin presented AI-generated images of herself during a City Council hearing, claiming that her team was able to create racy pictures using Grok and her official headshot within minutes. The images depicted her in revealing attire, with some prompts reportedly asking the AI to enhance features or add sexually suggestive elements.
Gabriela Castillo Madrid, SpaceX's senior manager of global government affairs, testified that Grok has "strict controls" to prevent the creation of such images, particularly those involving children or real individuals. However, when pressed by Menin, Madrid could not explain the specific instances presented.
"Madam speaker, I cannot speak to that specific instance," Madrid stated. "What I can say is that we do have strict controls, and I'm happy to get our safety team to look into that."
Menin's office stated that after asking Grok to place a banana in the images, the AI instead inserted a sex toy into the generated picture of the speaker. Menin expressed alarm, noting that these images were generated in the standard mode, not even the "spicy mode" that she claimed would horrify her daughter.
This incident occurred during a City Council hearing focused on the regulation of artificial intelligence. Earlier in October, the council heard testimony from executives at major AI companies, including Google, Meta, Anthropic, and OpenAI, regarding the potential dangers of advanced AI. Lawmakers are considering legislation that could mandate a "kill switch" or human override feature for AI systems operating within New York City.
SpaceX and its AI chatbot Grok have previously faced scrutiny, particularly internationally, over allegations of generating inappropriate or sexualized images without consent. A former SpaceX whistleblower also testified that he was told "tasteful nudity was acceptable" within the company's AI development, suggesting that the generation of such content might have been an intended, albeit controversial, outcome.