Elon Musk's xAI Refuses to Rein In Grok as Non-Consensual Deepfakes Run Wild
xAI’s chatbot Grok has enabled users on X (formerly Twitter) to easily create non-consensual deepfake images, including sexualized and nude portrayals of individuals without their permission. Prompts like “put her in a bikini” or “remove her clothes” can produce convincing altered images instantly. This has affected regular users and public figures alike, and has resulted in Grok generating sexualized images of minors, violating company policy and potentially U.S. law. Despite these incidents, some users exploit Grok for adult marketing and viral stunts, while others weaponize it for political messaging. Elon Musk has downplayed concerns, even sharing AI-generated bikini images of himself and others. Governments in France, India, and the UK have responded with investigations or announced bans. xAI claims it will improve safeguards, but moderation infrastructure is lacking since mass layoffs and the dissolution of Twitter's safety council. Grok’s “Spicy Mode” and minimal content restrictions reflect Musk’s commitment to free speech maximalism, even as harm and legal risks mount.
