Grok, the adaptable assistance tool from xAI for support, fact verification, and content development, is under the microscope by global regulators worried about its possible risks.

In January, X users found Grok producing sexualized images of minors in minimal clothing, categorized as child sexual abuse material (CSAM). xAI announced they are striving to improve the chatbot’s protective features.

Nonetheless, tackling this problem is not simple. Recent findings by Reuters, The Atlantic, and Wired have uncovered major weaknesses in the model’s protective measures, resulting in a rise in nonconsensual, sexual, and at times violent content crafted by Grok in response to user prompts. Investigations on X have unveiled a prevalent “undressing” problem, with numerous users’ public images being modified by the chatbot to show revealing attire. RAINN, a national anti-sexual assault organization, has deemed this behavior as AI or tech-facilitated sexual abuse.

Other AI chatbots and image creators have similarly been criticized for their insufficient safeguards against sexual content, including Meta’s AI companions.

In response to user apprehensions on X, CEO Elon Musk highlighted the gravity of producing unlawful content but contended that the legal accountability should rest with the users. Musk subsequently reacted to remarks from UK officials regarding a possible comprehensive ban, denouncing it as censorship and charging foreign governments with efforts to undermine free expression online.

Governments clamp down on xAI

Due to X’s teams’ delayed response, numerous nations are limiting access to xAI’s bot and standalone application while examining Grok’s protective measures, xAI’s actions, and possible breaches of online safety regulations.

The company has previously been subject to inquiries prior to the recent accusations of Grok facilitating CSAM.

Amid persistent scrutiny of X’s content moderation and algorithms, there are demands for