<\/p>\n
Hong Kong\u2019s privacy regulator has issued a warning against the potential misuse of X\u2019s Grok AI chatbot, as authorities worldwide grapple with controlling harmful and illegal content generated by advanced AI tools. The office is urging users to exercise caution and is taking steps to ensure compliance with privacy laws.<\/p>\n<\/div>\n
Hong Kong\u2019s privacy regulator has warned that X\u2019s AI chatbot Grok can be misused to produce indecent or malicious images and videos, and is contacting the company to seek explanations and further information, the Office of the Privacy Commissioner for Personal Data (PCPD) said in a statement. The development was reported by MLex and echoed by local press. [1]<\/a><\/sup>[2]<\/a><\/sup>[3]<\/a><\/sup><\/p>\n The PCPD cautioned that improper or malicious use of AI chatbots may breach Hong Kong\u2019s Personal Data (Privacy) Ordinance and \u201cmay also constitute other criminal offences\u201d, urging users not to share personal data without consent and to treat AI\u2011generated material with caution, according to the PCPD\u2019s media statement. The office has published guidance and resources aimed at helping users protect personal data when interacting with chatbots. [2]<\/a><\/sup>[1]<\/a><\/sup><\/p>\n The warning comes amid a growing patchwork of regulatory actions and safety alerts in other jurisdictions. Malaysia\u2019s regulator, the Malaysian Communications and Multimedia Commission, temporarily blocked access to Grok after finding repeated misuse to generate obscene, sexually explicit and non\u2011consensual manipulated images, including material involving women and minors; the block followed what the commission described as insufficient responses to formal notices sent to X Corp and xAI LLC. The Malaysian action has been reported by The Star and The Straits Times. [4]<\/a><\/sup>[7]<\/a><\/sup><\/p>\n International child\u2011safety groups have raised similar alarms. The UK\u2011based Internet Watch Foundation told The Guardian that users on a dark\u2011web forum boasted of using Grok \u201cImagine\u201d to create sexualised and topless images of girls aged 11 to 13, material the IWF said would meet the legal threshold for child sexual abuse material under UK law. That finding underlines concerns that image\u2011generation tools can be abused to create unlawful content extremely quickly. [5]<\/a><\/sup><\/p>\n xAI has said it has introduced technological restrictions on Grok\u2019s image editing features to prevent the alteration of images of real people into revealing clothing such as bikinis, a move reported by China Daily Hong Kong; the company claims these measures cover all users, including paid subscribers. The company\u2019s stated mitigations are a response to global backlash and regulatory pressure, but regulators in several jurisdictions have indicated they will monitor whether such controls are effective in practice. [6]<\/a><\/sup><\/p>\n The developments highlight how quickly AI tools are prompting national enforcement responses and legal scrutiny. According to MLex and the PCPD, regulators are focusing on both data\u2011privacy risks and criminal liability where AI output facilitates defamation, sexual exploitation or other offences. Industry data and watchdog reports suggest that regulators will expect clearer accountability from platform operators, faster remediation protocols and improved safety controls before restoring or permitting services in sensitive markets. [1]<\/a><\/sup>[2]<\/a><\/sup><\/p>\n For users and organisations, regulators recommend practical precautions: avoid uploading or sharing personal data without consent; review chatbots\u2019 privacy policies and terms of use; do not prompt systems to generate illegal content; and treat AI\u2011generated images and videos as potentially unverified. The PCPD has made resources available to help users assess privacy risks when using AI chatbots and has signalled it will take enforcement action if the law is breached. [2]<\/a><\/sup>[1]<\/a><\/sup><\/p>\n Taken together, the warnings, national blocks and vendor\u2011level mitigations illustrate an accelerating regulatory contest over generative AI governance: jurisdictions are testing both legal instruments and administrative measures to curb harms, while companies are rolling out technical fixes whose adequacy will be judged by independent watchdogs and regulators. Observers say the coming months will be decisive in determining whether ad hoc national steps coalesce into broader regulatory norms for image\u2011generation tools. [4]<\/a><\/sup>[5]<\/a><\/sup>[6]<\/a><\/sup>[1]<\/a><\/sup><\/p>\n ##Reference Map:<\/p>\n\ud83d\udccc Reference Map:<\/h3>\n
\n