Mr Cai Yinzhou asked the Minister for Digital Development and Information (a) whether IMDA will mandate content filters and digital watermarking for high-risk AI tools, following the blocking of Grok in Indonesia and Malaysia; and (b) how the Ministry intends to audit AI models accessible in Singapore for compliance with the Model AI Governance Framework.

Mrs Josephine Teo: Testing standards and technical solutions to manage artificial intelligence (AI)-related risks are still developing. Nonetheless, we have consistently emphasised the need for responsible AI development and deployment. We have sought to provide practical support through guidelines, like the Model AI Governance Framework, and testing tools, like AI Verify. We will update our policies as standards and technical safeguards mature.

Singaporeans are protected from online harms through safeguards such as the Codes of Practice for Online Safety and the upcoming Online Safety (Relief and Accountability) Act. The Member may refer to our previous response to Ms Cassandra Lee's Parliamentary Question on safeguards for users of Grok's app delivered at the 3 February 2026 Parliament Sitting. [Please refer to ​"Safeguards to Prevent Harm to Vulnerable Users of Gen AI Chatbot App", Official Report, 3 February 2026, Vol 96, Issue 15, Written Answers to Questions for Oral Answer not Answered by End of Question Time section.]

Link to Hansard: Link

Safeguards to Prevent Harm to Vulnerable Users of Gen AI Chatbot App

Ms Cassandra Lee asked the Minister for Digital Development and Information (a) what safeguards the Ministry has required of X's Grok app to prevent harm to vulnerable users, including children; (b) how the adequacy and enforceability of these safeguards are assessed before and during deployment; and (c) whether interim regulatory measures will be applied to limit exposure to or scale of use of Grok app, and if so, on what basis.

Mrs Josephine Teo: IMDA is engaging X on the recent issue of harmful and inappropriate content, such as non-consensual sexualised imagery, being generated and distributed on X using Grok. We note that X has taken some measures to address the matter globally, which includes preventing the Grok account on X from producing images of women and children in revealing clothing. 

There are safeguards in place to moderate Singapore users' exposure to harmful content on social media services and applications. The Codes of Practice for Online Safety require designated Social Media Services, including X, and designated app stores, to have in place measures to minimise users' exposure to harmful content, especially for children. This includes having in place community guidelines, content moderation measures and user reporting and resolution mechanisms. The designated services are also required to submit annual transparency reports detailing the measures they have put in place to combat harmful and inappropriate content.

In addition, the upcoming Online Safety (Relief and Accountability) Act will allow victims to seek redress if they face specified online harms, such as online harassment and intimate image abuse. The Online Safety Commission will be set up by June 2026.

We will continue to monitor the issue closely and will work with X to enhance online safety for Singapore users on its platform.

Link to Hansard: Link