Why Zambia’s Higher Education Must Embrace and Regulate AI Now
Technology is no longer knocking at the door of education. It has entered the classroom, the examination room, the library, the laboratory, the workplace and even the courts of law. The question before Zambia is no longer whether artificial intelligence should be allowed in education. That question has already been overtaken by reality. The real question is whether Zambia’s education system is ready to manage artificial intelligence before artificial intelligence begins to manage the education system.
This is the uncomfortable truth. Countries such as Zambia are still grappling with how to develop practical regulation on the use of AI in higher education, yet the world beyond has already moved into a new and more complicated phase. In some countries, the debate is no longer merely about students using ChatGPT to write assignments. The challenge has moved to AI-powered smart glasses, smart calculators, smart pens, hidden digital devices, real-time translation tools, voice assistants, automated writing platforms, coding tools, research generators and other technologies that are quietly changing the meaning of assessment, supervision, originality and competence.
Recent cases in countries such as South Korea and the United Kingdom show how smart devices are already creating new threats to examination integrity, with regulators warning about AI-enabled glasses, hidden earpieces and smart pens being used or feared in assessments. What should worry Zambia is not that these technologies exist elsewhere. What should worry us is that these technologies are portable, affordable, easily advertised online and capable of reaching our education system faster than our policies, laws and institutional controls.
We must therefore accept that higher education has entered a new era. This is the era of technology. This is the era of artificial intelligence. This is the era in which the credibility of education systems will not be judged only by the number of universities, colleges, lecturers, students and graduates, but by the ability of those systems to protect the integrity of learning in a technologically disruptive world.
Zambia has taken an important national step through the National Artificial Intelligence Strategy, which recognises AI as a tool for economic growth, improved public services and inclusive development. The strategy prioritises human capital development, digital infrastructure, data ecosystems, research and innovation, sectoral AI adoption, governance and ethical oversight. This is commendable. However, national strategy must now move quickly into sector-specific regulation, especially in education, health, finance, public service, agriculture, mining and justice. A strategy gives direction, but regulation gives discipline. Without discipline, technology can easily become a threat to public trust.
For higher education, this matter is urgent. The Higher Education Authority was established under the Higher Education Act No. 4 of 2013 with a mandate that includes quality assurance, registration of private higher education institutions, accreditation of learning programmes, advisory services and regulation of standards in higher education. In the age of AI, quality assurance can no longer remain limited to traditional questions of infrastructure, staffing, curricula, library holdings, student support and examination procedures. These remain important, but they are no longer enough. Quality assurance must now ask a new generation of questions.
How is AI being used in teaching and learning? How are institutions protecting academic integrity? How are lecturers redesigning assessments? Are students being trained on ethical AI use? Are research outputs being verified? Are institutional policies clear on what is permitted, restricted or prohibited? Are examination systems secure against smart technologies? Are graduates leaving university with genuine competence or merely polished submissions produced with technological assistance? These are no longer future questions. They are current questions.
The danger is not AI itself. The danger is unregulated AI. AI can support learning, improve research, assist students with disabilities, enhance data analysis, support language development, improve administration and expand access to knowledge. Properly used, AI can strengthen education outcomes. It can help lecturers prepare better learning materials, assist students to understand complex concepts, support institutional planning and improve efficiency in quality assurance processes.
But the same technology, when abused, can weaken the very foundation of education. It can produce assignments that students do not understand. It can generate research proposals without genuine inquiry. It can create legal arguments without legal reasoning. It can influence medical diagnosis without sufficient human judgement. It can produce policy briefs without policy understanding. It can help a student pass an examination without possessing the competence that the grade claims to certify.
This is why the assessment question has become the most serious policy question in higher education today. For many years, education systems relied heavily on written assignments, take-home essays, examinations and research projects as evidence of learning. AI has not destroyed these tools, but it has made them insufficient on their own. A well-written assignment is no longer automatic evidence that a student has mastered the subject. A polished research paper is no longer automatic evidence of independent thinking. A high grade is no longer automatic evidence of competence unless the assessment system has been designed to verify actual learning.
This is why Zambia’s transition towards competence-based education must be taken seriously. Competence-based education cannot be reduced to changing curriculum language or inserting the word “competence” in programme documents. It must change how we teach, how we assess, how we supervise, how we moderate, how we verify skills and how we certify graduates. In the age of AI, competence must be demonstrated, not merely declared.
Universities and colleges must begin to combine traditional assessment with practical demonstrations, oral defence, viva voce examinations, workplace-based learning, simulations, laboratory-based tasks, portfolios, reflective journals, supervised writing, project presentations, community-based assignments and real-time problem-solving exercises. Where AI is permitted, students must be required to disclose how it was used. Where AI is prohibited, institutions must make that prohibition clear and enforceable. Where AI is encouraged, it must be aligned with learning outcomes and ethical standards.
The worst response Zambia can make is to ban AI blindly. That would be both unrealistic and intellectually lazy. Students, lecturers, researchers and professionals are already using AI in one way or another. Banning it without teaching ethical use will only drive the practice underground. The better response is to classify AI use according to purpose. There must be assessments where AI use is not allowed at all. There must be assessments where limited AI assistance is allowed with disclosure. There must be assessments where AI is deliberately integrated because the learning outcome requires students to demonstrate digital fluency, critical judgement and responsible use of technology.
This is the balance Zambia must pursue: not fear of AI, but governance of AI; not resistance to technology, but intelligent regulation of technology; not blind adoption, but purposeful assimilation.
The second danger is inequality. If AI becomes a hidden advantage for students with better devices, better internet access and better exposure, then technology may widen the gap between privileged and disadvantaged learners. Zambia must therefore approach AI as both a quality issue and an equity issue. Rural students, students in under-resourced institutions, students with limited internet access and students from poor households must not be left behind. AI literacy must become part of the national education conversation, not a privilege for urban institutions.
The third danger is institutional unreadiness. Many higher education institutions are still developing basic internal quality assurance systems, yet they are now being confronted by AI-driven disruption. Some institutions may not have AI policies. Some lecturers may not know how to detect AI misuse. Some examination regulations may still be written for the era of mobile phones and paper notes, not smart glasses and generative systems. Some research ethics committees may not yet have guidance on AI-generated data, AI-assisted writing or AI-supported analysis. This gap between technological reality and institutional preparedness is where quality risk begins.
The fourth danger is employer distrust. If employers begin to doubt the competence of graduates, the reputation of higher education suffers. A qualification is not merely a certificate. It is a public promise that the graduate has achieved a certain level of knowledge, skill, judgement and professional readiness. When technology makes it possible for students to appear competent without being competent, the burden falls on quality assurance systems to protect the value of qualifications.
This is why AI regulation in higher education must not be treated as an administrative matter. It is a national development matter. Zambia cannot build a 21st-century economy with 20th-century assessment systems. We cannot speak of innovation while leaving universities and colleges without guidance on AI. We cannot demand competent graduates while allowing assessment systems to remain vulnerable to technologies they were never designed to manage.
The Higher Education Authority, working with the Ministry of Education, the Ministry of Technology and Science, universities, colleges, professional bodies, employers, student representatives and other stakeholders, must help lead the development of a higher education AI regulatory framework. Such a framework should address ethical AI use, academic integrity, assessment design, research conduct, data privacy, institutional policy requirements, digital infrastructure, staff training, student orientation and quality assurance standards.
Every higher education institution should be required to develop an institutional AI policy. Every policy should indicate how AI affects teaching, learning and assessment. Every lecturer should be trained in AI-aware pedagogy. Every student should be oriented on acceptable and unacceptable AI use. Every research policy should address AI-assisted research. Every examination regulation should be updated to deal with smart technologies. Every quality assurance audit should ask whether the institution is prepared for the AI era. This is not optional. It is the new frontier of quality assurance.
UNESCO has already called on countries and institutions to guide the use of generative AI in education and research through policies that protect human agency, inclusion, equity, ethics and pedagogical value. Zambia must not wait until the problem becomes a scandal. We must not wait until employers begin to reject graduates. We must not wait until courts, hospitals, schools, engineering firms and public institutions begin to question whether professional outputs are genuinely human, ethically assisted or irresponsibly generated. We must act now.
The country must also invest in digital infrastructure. It is not enough to regulate AI if institutions lack reliable internet, electricity, cybersecurity systems, digital libraries, data protection capacity and trained personnel. Zambia’s own digital economy context shows that infrastructure, electricity reliability and digital skills remain important barriers to technological development. Therefore, AI regulation must move together with AI capacity building. Regulation without capacity becomes punishment. Capacity without regulation becomes chaos. Zambia needs both.
At the same time, we must avoid a narrow view of AI as a cheating problem only. That would be a mistake. AI is also a productivity tool, a research tool, a teaching tool, a planning tool, a health tool, an agricultural tool, a mining tool, a governance tool and an innovation tool. The purpose of regulation is not to suffocate innovation. The purpose of regulation is to ensure that innovation serves national development without destroying trust, fairness, accountability and competence.
The future will belong to countries that can do two things at once: adopt technology quickly and regulate it wisely. Countries that adopt without regulation will face abuse. Countries that regulate without adoption will fall behind. Countries that do neither will be left behind completely.
For Zambia, the choice is clear. We must accelerate technological assimilation, but we must match it with equally rapid regulation. We must prepare our students to use AI, but also prepare them to think beyond AI. We must protect assessment integrity, but also redesign assessment for a world in which information is abundant and competence is what truly matters. We must empower lecturers, but also demand institutional accountability. We must encourage innovation, but never at the expense of educational credibility.
The 21st century will not be kind to education systems that move slowly. Technology will not wait for policy meetings, committee reports and delayed implementation. It will continue to evolve. It will continue to enter our classrooms. It will continue to challenge our examinations. It will continue to test our understanding of knowledge, competence and integrity.
Therefore, Zambia must decide whether to be a passive consumer of technology or an intelligent governor of technology. The higher education sector must be placed at the centre of this national decision because universities and colleges produce the professionals who will run the country’s hospitals, courts, schools, mines, farms, laboratories, businesses and public institutions.
If higher education fails to respond, the whole country becomes vulnerable. But if higher education responds with wisdom, courage and speed, Zambia can turn AI from a threat into a national advantage.
This is the moment to act. The era of AI is here. The regulation of AI in higher education cannot wait.
By Dr. Martin Mushumba
Public Policy and Education Quality Assurance Expert
