A preteen girl sits at a desk using an AI chatbot to solve her homework.
Finato Studio | E+ | Getty Images
The UK government has made AI chatbots subject to requirements to combat illegal content, closing a “loophole” in new online safety laws that could lead to them being fined or blocked.
After the country’s government harshly criticized Elon Musk’s X over sexually explicit content created by chatbot Grok, Prime Minister Keir Starmer announced new measures meant for chatbots such as OpenAI’s ChatGPT. Google with gemini microsoft Co-pilots are expected to be included in the government’s online safety legislation.
Platforms will be required to comply with “illegal content obligations” or “face the consequences of violating the law,” the announcement said.
This comes after the European Commission investigated MuskX in January for spreading sexually explicit images of children and other individuals. Mr Starmer led calls for Mr Musk to stop this.
British Prime Minister Keir Starmer during a press conference on Monday, January 19, 2026 in London, England.
Bloomberg | Bloomberg | Getty Images
Earlier, British media watchdog Ofcom launched an investigation into X for allegedly disseminating sexually explicit images of children and other individuals.
“The actions we have taken against Grok send a clear message that no platform will be given a free pass,” Starmer said in announcing the latest action. “We are closing loopholes that put children at risk and laying the foundations for further action.”
Starmer spoke on Monday about new powers, including setting minimum age limits for social media platforms, restricting harmful features such as infinite scrolling, and restricting children’s use of AI chatbots and access to VPNs.
One of the measures announced would force social media companies to retain data after a child’s death, unless their online activity was clearly unrelated to the death.
“We are acting to protect the health of children and help parents navigate the minefield of social media,” Starmer said.
Alex Brown, head of TMT at law firm Simmons & Simmons, said the announcement shows how governments are taking a different approach to regulating rapidly evolving technologies.
“Historically, our lawmakers have been reluctant to regulate this technology, and instead sought to regulate its uses, and for good reason,” Brown said in a statement to CNBC.
He said regulations that focus on specific technologies risk becoming outdated quickly and losing aspects of their use. Brown said generative AI exposes the limitations of online safety laws that focus on “regulating services, not technology.”
He said Mr Starmer’s latest announcements showed the UK government wanted to address the risks “not only arising from user-generated content and the functionality of the platforms, but also from the design and operation of the technology itself”.
Children and adolescents’ access to social media has come under increased scrutiny in recent months, with lawmakers citing harm to their mental health and well-being. In December, Australia became the first country to enact a law banning social media use by under-16s.
Australia’s ban requires apps such as Alphabet’s YouTube, Meta’s Instagram and ByteDance’s TikTok to require age verification methods, such as uploading an ID or bank account information, to prevent anyone under 16 from creating an account.
Spain became the first European country to implement a ban earlier this month, and France, Greece, Italy, Denmark and Finland are also considering similar proposals.
In January, the UK government launched a consultation on banning social media for under-16s.
Additionally, the House of Lords, the country’s unelected upper legislative chamber, passed amendments to the Child Welfare and Schools Bill last month that include a ban on social media for under-16s.
The next step will be for the bill to be debated in the lower house of parliament. Both chambers must agree on any changes before the bill can be passed.
