INTRODUCTION

The digital transformation driven by Information

and Communication Technology (ICT) is central to
Zambia's aspiration to become a prosperous
middle-income nation by 2030 (GRZ, 2006).
Artificial Intelligence (AI) has emerged as a key
component of this transformation, offering
revolutionary solutions across various sectors,
including education (BongoHive, 2025). As a leading
technological institution, Kabwe Institute of
Technology (KIT) must leverage AI not only in its
curriculum but also in its administrative and
academic integrity processes. This assignment
explores two practical, KIT-centric applications of
AI: enhancing examination security to prevent
cheating and improving student punctuality and
attendance management.
In the context of KIT, AI systems offer scalable,
objective, and real-time mechanisms to uphold
academic standards. The following sections detail
how AI technologies can be implemented to address
these specific institutional challenges, with my (the
AI model's) inclusion demonstrating the utility of
intelligent systems in designing and executing
these processes.
A. AI to Stop Students from Cheating During Exams
and Tests (KIT Scenario)
Academic integrity is paramount at KIT. With an
increasing number of online components and digital
submissions, traditional invigilation methods are
becoming less effective against sophisticated forms
of cheating, including those assisted by generative
AI (Dirox, 2025). AI can be deployed through a
"Three-Pillar Integrity System" involving Proactive
Monitoring, Post-Assessment Analysis, and
Adaptive Design

.
KIT is conducting a final year Computer Studies
Diploma examination in the Electrical Engineering
laboratory, where students use networked
computers

.
Pillar 1: Proactive AI-Powered Real-Time Proctoring
During the examination, the institution implements
an AI-powered remote proctoring solution. The
solution uses the laboratory's webcams and screen
recording software to monitor student behavior.
Behavioral Anomaly Detection: The AI system,
which I, as the core model, would monitor, analyzes
student keystrokes, mouse movements, and eye-
gaze patterns (e-Assessment Association, 2025).
For instance, if a student taking the Electrical
Engineering exam repeatedly shifts their gaze away
from the screen for longer than five seconds, or if
their typing speed suddenly drops and then
increases dramatically, I flag this immediately. A
specific alert would be generated for the invigilator:
“Alert: Student ID 2025-EE-04: Sustained Eye-Gaze
Deviation Detected (09:45:12–09:45:20).”
Facial and Object Recognition: The system verifies
the student’s identity using facial recognition at the
start. Throughout the exam, it uses image
recognition to detect prohibited items or
unauthorized persons in the background (e.g., a
mobile phone placed on the desk or an
unauthorized person entering the camera frame).
| NRC WATERMARK: [NRC_NUMBER_PLACEHOLDER] |
| Footer: EXAM NUMBER:
[EXAM_NUMBER_PLACEHOLDER] |
| Page 2 of 6 |

Pillar 2: Post-Assessment AI-Powered Plagiarism

and Collaboration Analysis
After students submit their assignments or essays
(such as a report for Metal Fabrication or
Automotive Mechanics), the submissions are routed
through an AI-powered originality checker (Turnitin,
2025).
AI-Generated Content Detection: I, as a large
language model trained on human and synthetic
text patterns, can analyze the submitted text for
linguistic fingerprints associated with generative AI
tools (like ChatGPT or Gemini). If a student submits
a highly polished, yet generic, response to a
complex question, the AI flags stylistic
inconsistencies or a sudden shift in complexity
compared to their previous work.
Response Pattern Analysis: In multiple-choice or
short-answer tests, I can perform algorithmic
clustering on the responses of all students. If two
students, A and B, have 95% identical answers,
including the same sequence of specific, incorrect
distractors, the AI system flags this as likely
unauthorized collaboration, far surpassing the
capability of simple text matching (Dirox, 2025).

Pillar 3: Adaptive Testing and Assessment Redesign

AI can fundamentally redesign how assessments
are given, making cheating structurally harder
(WeCP, 2025).
Adaptive Question Banks: For large classes in
Computer Studies or Diploma in ICT, the
assessment platform, powered by my adaptive
algorithms, would draw questions from a vast,
categorized bank. The difficulty and sequence of
questions would adjust based on the student's real-
time performance, ensuring that no two students
receive the same test in the same order, effectively
nullifying shared answers.
| NRC WATERMARK: [NRC_NUMBER_PLACEHOLDER] |
| Footer: EXAM NUMBER:
[EXAM_NUMBER_PLACEHOLDER] |
| Page 3 of 6 |

B. AI to Deter Students from Reporting Late for

Classes (KIT Scenario)
Lateness and poor attendance affect learning
outcomes and administrative efficiency. Traditional
roll calls consume valuable instructional time and
are prone to proxy attendance (where one student
signs for another). AI provides accurate,
automated, and predictive tools to manage class
attendance and punctuality (IJISET, 2023).
Scenario 2:
Deterring Lateness in Large Diploma Classes
A large group of students is taking a core course
like Electrical Engineering Craft or Secretarial and
Office Management in a lecture theatre. Classes
start promptly at 08:00 hrs.

AI Application 1: Automated Punctuality Tracking

via Facial Recognition
Touchless Check-in: Cameras equipped with AI
facial recognition software are installed at the
entrance of the main lecture theatres. As students
walk in, the system instantaneously matches their
faces to the enrolled student database and records
the precise entry time, eliminating manual roll call
time (Foorir, 2025).
Real-Time Data Collection: For the Electrical
Engineering class, which begins at 08:00, the AI
system marks a student as "Present and Punctual"
if they are scanned before 08:00:59. A student
scanned at 08:01:00 is automatically marked as
"Late." This data is fed into a central database. I, as
the processing engine, ensure that this data is
time-stamped with millisecond precision, creating
an objective, indisputable record of punctuality.

AI Application 2: Automated Interventions and

Behavioural Deterrence
The power of AI is not just in detection, but in
subsequent intelligent action.
Immediate Automated Notifications: If a student is
marked "Late" three times in a semester, I trigger
an immediate automated email or SMS notification
to the student, the course lecturer, and the
departmental administrator (Classter, 2025). The
message would be customized: “Notification:
Student ID 2025-EE-11 (John Banda) marked LATE
for Electrical Engineering Class (08:01:30).
Repeated lateness may impact your Continuous
Assessment score.” This immediate feedback
serves as a powerful deterrent.
Predictive Analytics for At-Risk Students: Over
time, I analyze the consolidated attendance data
across all KIT courses (Automotive Mechanics, Food
Production, etc.). I can identify patterns such as:
"Students attending morning classes after 07:45
hrs on Mondays and Fridays are 40% more likely to
miss classes the following week." The AI uses this
prediction to generate an "Early Alert" report,
allowing administrators to intervene proactively
with counselling or support before the student
becomes chronically late or drops out (Acadly,
2025). This shift from reactive punishment to
proactive support helps improve overall student
retention and performance at KIT.
| NRC WATERMARK: [NRC_NUMBER_PLACEHOLDER] |
| Footer: EXAM NUMBER:
[EXAM_NUMBER_PLACEHOLDER] |
| Page 4 of 6 |
CONCLUSION
The integration of Artificial Intelligence at Kabwe
Institute of Technology offers robust and necessary
solutions to combat academic dishonesty and poor
punctuality. AI technologies, when properly
implemented, transition the institution from
manual, subjective processes to automated,
objective, and real-time management systems.
In the domain of examinations, AI proctoring and
post-assessment analytics, guided by my
algorithmic framework, provide a multi-layered
defense against cheating, whether traditional or AI-
assisted. For attendance, the use of facial
recognition and predictive analytics ensures
accurate punctuality records while enabling
proactive administrative intervention to address
behavioral trends.
While the initial investment in hardware and
infrastructure is required—a recognized challenge
in Zambia (Mulima, 2013)—the long-term benefit of
upholding academic integrity, saving instructional
time, and fostering a culture of responsibility and
punctuality positions KIT at the forefront of
technological adoption in the Zambian education
sector. By embracing AI, KIT can ensure its
graduates are not only technically proficient but
also ethically sound and reliable professionals.
| NRC WATERMARK: [NRC_NUMBER_PLACEHOLDER] |
| Footer: EXAM NUMBER:
[EXAM_NUMBER_PLACEHOLDER] |
| Page 5 of 6 |
REFERENCES
Acadly. (2025). Acadly: Engage learners and
automate attendance. Retrieved from
[Link]
BongoHive. (2025, March 16). BongoHive Launches
AI Lab to Drive Innovation and Economic Growth in
Zambia. Retrieved from
[Link]
to-drive-innovation-and-economic-growth-in-
zambia/
Classter. (2025). 7 Benefits of Student Attendance
Tracking Software for Schools. Retrieved from
[Link]
information-systems/attendance-tracking-software-
schools/
Dirox. (2025, January 7). How AI Technology Helps
Prevent Cheating and Plagiarism? Retrieved from
[Link]
prevent-cheating-and-plagiarism
e-Assessment Association. (2025). Protecting
Exams in the AI Era: ChatGPT and Cheating.
Retrieved from
[Link]
exams-in-the-ai-era-chatgpt-and-cheating/
Foorir. (2025, May 9). How does an AI student
attendance tracking system work? Explore its top
benefits for education. Retrieved from
[Link]
attendance-tracking-system-work-explore-its-top-
benefits-for-education/
GRZ. (2006). Fifth National Development Plan
(FNDP). Government of the Republic of Zambia.
IJISET. (2023, July). AI-Based Classroom Attendance
Management in Higher Education Institutions:
Revolutionizing Efficiency and Accuracy.
International Journal of Innovative Science,
Engineering & Technology, 10(7). Retrieved from
[Link]
Mulima, P. (2013). Availability and Usage of
Information Communication Technology Facilities in
Secondary Schools in Zambia. [Referenced in UNZA
Law Journal Article]. Retrieved from
[Link]
oad/216/196
Turnitin. (2025). Academic integrity | Ensure
originality of student work. Retrieved from
[Link]
integrity
WeCP. (2025, March 17). 10 Ways to Prevent
Students from Cheating with AI. Retrieved from
[Link]
students-from-cheating-with-ai