OpenAI said on November 12 that it pledged US$2 million to fund tools that measure AI’s risks and opportunities for children.
The money goes to the research network under the Coalition for Children’s Rights and Protection in the Age of AI.
In Kenya, the Communications Authority plans compliance checks on firms that serve children online, so the same question reaches telcos, platforms and parents here.
OpenAI funds measurement tools for researchers
OpenAI signed the Charter of Engagement for Safer AI for Children at the Paris Peace Forum earlier in the week. French President Emmanuel Macron launched the charter, and Microsoft also signed it.
OpenAI said its contribution pays for shared measurement work: tools researchers can use to assess AI’s risks and opportunities for children. The company said the findings should inform how products are built and how policy is written.
The charter sets principles for AI matched to a child’s age. It also sets safeguards for research independence, including publication of negative or inconclusive findings.
The France backed coalition includes UNICEF, the United Nations Children’s Fund, and centres children’s rights in AI governance. OpenAI’s usage policies already prohibit child exploitation.
Governments signed a separate declaration in Paris
The UAE, Australia, Egypt, South Africa, Spain, Germany and the United Nations signed the Children First Declaration promoted by Macron. Macron said no AI should reach children before it is tested, evaluated and authorised.
AI developers including OpenAI and Anthropic also pledged €8 million, or $8.9 million, for research and tools on how AI affects child development. The European Union’s AI Act takes a risk-based approach, and child rights group 5Rightspoints to provisions on assessing and mitigating risks to children.
Communications Authority guidelines cover Kenyan telcos and platforms
The Communications Authority of Kenya adopted its Industry Guidelines for Child Online Protection and Safety on April 29, 2025, and they took effect on October 29, 2025. Khusoko reported the launch in April 2025.
The guidelines require service providers to publish corporate child safety policies, name a responsible officer and build safety into ICT products. They also urge industry to use age verification and to classify content that does not suit all subscribers.
ICT Cabinet Secretary William Kabogo said the ministry, through the Authority, plans compliance checks against the guidelines. He told the Senate on May 13, 2026 that the first stage is audits across the industry. He said results from a National Survey on Child Online Protection and Safety will inform future regulation..
Kabogo also warned that safety measures built for English may fail for children who use other languages, Eastleigh Voice reported. The outlet said Kenya joined Spain, the European Union and France in launching a coalition on children and AI.
AI child safety rules in Kenya rest on guidelines and scattered laws
KICTANet’s Policy Brief No. 38, researched by Dr Owiso Owiso and published in March 2026, drew on a Nairobi roundtable held on November 11, 2025. It says children and youth make up about 75 percent of Kenya’s population, and that 26 percent of rural and 81.6 percent of urban households have internet access.
Section 33 of the Data Protection Act bars processing a child’s personal data without parental or guardian consent and requires age verification. The brief says the Authority’s guidelines and the Data Protection Commissioner’s 2025 guidance notes carry no binding force of law. It also says criminal law does not reach online platforms as intermediaries, and that at least eight agencies hold overlapping mandates.
The brief recommends binding rules on parental consent and age verification, criminal liability for intermediaries and binding safety by design duties. It advises platforms to collect age ranges rather than exact birth dates, and to apply stricter checks to adult content and financial services.
A KICTANet roundtable report dated September 24, 2026 adds that Kenya has no single law on age verification or parental consent. Participants cited Australia’s social media ban for under sixteens as a case where regulator data and research show most affected users found ways around it.
What changes for firms and families
Firms holding a Communications Authority licence face audits against the 2025 guidelines. Parents who open social media accounts for children fall under the Data Protection Act’s consent rule, and the Commissioner’s guidance notes recommend parental consent for those accounts.
The OpenAI money funds measurement tools, not enforcement. Kenyan regulators have not said whether they will use the coalition’s research.
