California Builds AI Oversight Unit, Presses XAI Investigation

California is moving to tighten regulation of artificial intelligence, announcing plans to create a dedicated AI oversight unit while continuing an investigation into the AI company xAI, according to state officials.

Attorney General Rob Bonta said his office will establish a specialised AI regulation and enforcement team within the California Department of Justice. The unit will focus on monitoring AI systems, enforcing existing consumer protection laws, and addressing potential harms linked to rapidly expanding AI use.

The announcement comes as California presses its xAI investigation over concerns that its chatbot, Grok, may have generated sexually explicit images without consent. State officials are examining whether the system produced such content in violation of California law, including material that could involve minors.

• California sets up dedicated AI oversight unit amid xAI probe
• State presses investigation into xAI over Grok image generation concerns
• California steps in as federal action on AI regulation lags

Bonta said his office issued a cease-and-desist letter to xAI last year and is seeking confirmation that the company has fully stopped the disputed practices. He added that investigators remain concerned about whether safeguards are sufficient and consistently applied.

xAI has said it has introduced technical measures to block certain image generation requests and has argued that it complies with applicable laws. However, California officials say they are not satisfied that all risks have been addressed, and the inquiry remains active.

The move highlights California’s effort to step in as federal lawmakers debate AI governance. Bonta said Congress has been slow to act on technology oversight in the past, including on data privacy, leaving states to respond to emerging risks.

California lawmakers are also considering legislation that would formally require the attorney general’s office to maintain in-house AI accountability and enforcement capacity. Supporters argue that state-level action is necessary as AI systems become more widely used across business, media and government.

Also Read: Tarique Rahman Takes Oath as Bangladesh PM After Landslide Win

Connecticut Attorney General William Tong has backed California’s approach, warning that unchecked AI and social media harms could create large-scale consumer risks if regulators fail to act early.