Kenya’s push to embrace Artificial Intelligence (AI) must be matched by stronger parliamentary oversight, accountability and protection of citizens’ rights.
Chairperson of the Communications, Information and Innovation Committee (CII), John Kiarie said Parliament must ensure that technological innovation does not move faster than the laws, institutions and safeguards required to protect Kenyans from the potential risks associated with AI.
Kiarie who is Dagoretti South MP said Parliament has a critical role in scrutinising how AI technologies are developed, purchased and deployed, particularly when automated systems are used to make or influence decisions affecting citizens.
“Parliament must carefully examine how AI is being developed, procured and deployed, particularly where automated systems can influence decisions affecting people’s livelihoods and access to essential services,” Kiarie said.
He was speaking during a webinar titled “Synchronizing AI Innovations: Legislative Interventions Towards Harnessing AI in Africa,” convened by the Centre for Parliamentary Studies and Training (CPST) and moderated by Parliamentary Joint Services Director General Clement Nyandiere.
The forum brought together legislators, government officials and technology-sector stakeholders to examine how Kenya can take advantage of AI while managing risks associated with its growing use.
Keep Reading
- Comfort and Consequences: The Rise of AI Confidants
- Beyond social media: When governments regulate the wrong thing
- How AI could help fix Kenya's overstretched healthcare system
Kiarie’s committee is expected to play a central role in examining AI policy, including how government agencies procure AI systems, whether public funds deliver value and whether citizens’ rights are adequately protected.
The proposed roadmap includes identifying existing AI systems and pilots, introducing interim rules for AI use and procurement, designating accountable officers and developing risk classifications and impact assessments.
High-risk systems could subsequently face certification, interoperability testing and publication of performance and audit findings.
Kiarie said Kenya’s objective should not be to slow technological advancement but to ensure that innovation remains responsible and people-centred.
“The central challenge, he said, is determining how Kenya can harness AI while safeguarding human dignity, equality, privacy, freedom of expression, access to information and accountability,” he said.
Participants noted that AI is likely to transform sectors including employment, education, healthcare, financial services, social protection, security and justice.
Stakeholders questioned whether Kenya’s existing legal framework is adequate to regulate rapidly developing AI technologies or whether additional legislation and safeguards will be necessary.
Participants warned that systems trained using historical data containing discrimination, exclusion or under-representation could produce biased outcomes in areas such as recruitment, lending, education, healthcare, insurance and access to government services.
Dr Grace Githaiga, Convenor of the Kenya ICT Action Network (KICTANet), said Kenya should not have to choose between innovation and protection of human rights.
She warned that AI was changing the nature of privacy and data-protection concerns because systems can analyse vast amounts of information, generate profiles and make inferences about individuals.
“The concern is no longer simply about what personal information is collected and stored,” Githaiga said.
She said policymakers must examine the data used to train AI models, what systems can infer about individuals and how those inferences could affect people’s lives.
ICT Authority Chief Executive Officer Jessy Maruti said government agencies must remain accountable when using AI, arguing that responsibility cannot be transferred to technology providers or algorithms.
“Government accountability cannot be delegated to a vendor or an algorithm,” Maruti said.
He identified infrastructure, standards, institutional capacity and accountability as key pillars for responsible AI adoption.
Maruti also stressed that digital transformation should extend beyond internet connectivity to include reliable electricity, fibre networks, cloud infrastructure, secure computing capacity, trusted data systems, cybersecurity and digital skills.
Although mobile broadband coverage reaches an estimated 96 per cent of Kenyans, participants noted that a significant usage gap remains, meaning network availability does not automatically translate into meaningful digital inclusion.
Another major recommendation was the need for human oversight whenever AI is used in decisions affecting citizens’ rights, safety, benefits or access to public services.
Maruti called for mechanisms allowing citizens to challenge AI-assisted decisions, request human review and obtain remedies when automated systems make errors.
He also advocated a risk-based regulatory framework, where low-risk AI applications face basic safeguards while high-risk systems are subjected to stronger controls such as impact assessments, independent reviews, external validation, registration and regular audits.