xAI places limits on Grok picture modifying after considerations from California, Europe

[ad_1] Merchandise 1 of two Elon Musk speaks throughout a press convention with U.S. President Donald Trump (not pictured), on the White Home in Washington, D.C., U.S., Might 30, 2025. REUTERS/Nathan Howard

[1/2] Elon Musk speaks throughout a press convention with U.S. President Donald Trump (not pictured), on the White Home in Washington, D.C., U.S., Might 30, 2025. REUTERS/Nathan Howard <a data-testid="Link" target="_blank" referrerpolicy="no-referrer-when-downgrade" href="https://www.reutersconnect.com/item/us-president-donald-trump-and-elon-musk-attend-a-press-conference-at-the-white-house/dGFnOnJldXRlcnMuY29tLDIwMjU6bmV3c21sX1JDMklTRUE5UkdNMQ%3D%3D/?utm_medium=rcom-article-media&utm_campaign=rcom-rcp-lead" rel="noopener" class="text-module__text__0GDob text-module__inherit-color__PhuPF text-module__inherit-font__1P1hv text-module__inherit-size__EyiQW link-module__link__INqxZ link-module__underline_default__-okuC link-module__with-icon__qlg76 collapsible-caption-module__license__CPQvV">Purchase Licensing Rights , opens new tab xAI blocks all customers from modifying pictures of 'actual folks in revealing clothes' California officers demand solutions from xAI after its Grok chatbot generated sexualized photos of girls, minors Legal professional Basic says the state desires X to 'cease the unfold of this content material' Musk says he's unaware of 'bare underage pictures' generated by Grok Jan 14 (Reuters) - Elon Musk's synthetic intelligence firm xAI stated late on Wednesday it imposed restrictions on all customers of its Grok AI chatbot that restrict picture modifying after the service produced sexualized pictures that sparked considerations amongst world regulators. "We've applied technological measures to forestall the Grok account from permitting the modifying of pictures of actual folks in revealing clothes equivalent to bikinis. This restriction applies to all customers, together with paid subscribers," the corporate stated in an X put up.

Join <a data-testid="Link" referrerpolicy="no-referrer-when-downgrade" href="https://www.reuters.com/world/asia-pacific/musk-says-unaware-grok-generating-explicit-images-minors-2026-01-14/undefined?location=article-paragraph&redirectUrl=%2Fworld%2Fasia-pacific%2Fmusk-says-unaware-grok-generating-explicit-images-minors-2026-01-14%2F" class="text-module__text__0GDob text-module__inherit-color__PhuPF text-module__inherit-font__1P1hv text-module__inherit-size__EyiQW link-module__link__INqxZ link-module__underline_default__-okuC">here.

Hyper-realistic pictures of girls manipulated to seem like they have been in microscopic bikinis, in degrading poses or lined in bruises started flooding social media platform X this month. In some circumstances, minors have been digitally stripped all the way down to swimwear, sparking broad criticism.

Grok final week started permitting solely paying subscribers to make use of its picture technology and modifying options. X final week curtailed Grok's skill to generate or edit pictures publicly for a lot of of its customers, however the chatbot nonetheless privately produced sexually charged pictures on demand on Wednesday earlier than xAI's announcement, Reuters discovered.

Billionaire Musk owns xAI, which in flip owns X, previously referred to as Twitter.

xAI added on Wednesday that it blocks customers based mostly on their location from producing pictures of individuals in skimpy apparel in "jurisdictions the place it is unlawful". It didn't identify these jurisdictions.

CALIFORNIA OFFICIALS DEMAND ANSWERS California's governor and legal professional normal stated earlier on Wednesday they have been demanding solutions from xAI after Musk stated he was not conscious of any "bare underage pictures" generated by Grok.

"We’re demanding instant solutions from xAI on their plan to cease the creation & unfold of this content material," California Legal professional Basic Rob Bonta wrote on X. Governor Gavin Newsom referred to as on Bonta "to right away examine the corporate and maintain xAI accountable."

The feedback from Newsom and Bonta have been essentially the most severe to this point by U.S. officers addressing the explosion of AI-generated nonconsensual sexualized imagery on X. The California transfer added to stress Musk is dealing with within the U.S. and world wide. Lawmakers and advocacy groups have referred to as for Apple (AAPL.O) , opens new tab and Google (GOOGL.O) , opens new tab to drop Grok from app shops. Authorities officers have threatened action in Europe and the United Kingdom . Indonesia temporarily blocked entry to Grok. At first, Musk publicly laughed off the controversy, posting humorous emojis in response to different customers' feedback concerning the inflow of sexualized images. Extra just lately, X has stated it treats reviews of kid sexual abuse materials severely and polices it vigorously.

Musk stated earlier on Wednesday he was "not conscious of any bare underage pictures generated by Grok. Actually zero."

X didn't instantly reply to questions concerning the California announcement and Musk's feedback.

xAI didn't reply on to an emailed request for touch upon California officers' statements or Musk's put up that he was unaware of sexualized imagery of minors. Reuters acquired its generic autoreply message for inquiries: "Legacy Media Lies."

Reporting by Raphael Satter in Washington, Akash Sriram in Bengaluru and Chris Thomas in Mexico Metropolis; Modifying by David Gregorio, Cythia Osterman and Jamie Freed

Our Requirements: The Thomson Reuters Trust Principles. , opens new tab

Purchase Licensing Rights Reporter protecting cybersecurity, surveillance, and disinformation for Reuters. Work has included investigations into state-sponsored espionage, deepfake-driven propaganda, and mercenary hacking.