evidoria

← Back to browse

Good practice Imported

KovsieChat — University of the Free State's AI Chatbot Scales Student Support in Bloemfontein

South Africa · Bloemfontein · See the South Africa profile

Top 81% 33/100 · Ask Evidence Copilot about this practice

The University of the Free State grew a COVID-era chatbot into KovsieChat, an AI assistant for admissions, registration and financial-aid queries at its ~45,000-student multi-campus university, easing staff workload during peak enrolment.

KovsieChat — University of the Free State's AI Chatbot Scales Student Support in Bloemfontein

Details

Promoter
University of the Free State
Period
2020–2026
Keywords
higher education, student administration, chatbot, enrolment support

Description

The University of the Free State (UFS) in Bloemfontein, South Africa, first built a simple rules-based chatbot, Kovsie2Bot, during the COVID-19 pandemic to keep prospective and current students informed when in-person channels were disrupted. It was later rebranded KovsieChat and, in May 2024, upgraded to a generative-AI assistant built on the EduBot platform (developed by SIS Global and its AI division, Verge AI) running on Microsoft Azure OpenAI's GPT-4.

KovsieChat operates 24/7 and answers questions on admission requirements, course and faculty information, entry scores, registration steps, scholarships and campus life. It supports South Africa's official languages plus additional languages, and includes voice-note input to widen access for students with limited typing literacy or connectivity.

According to the deploying vendor's own case study, the GPT-4 upgrade helped the bot process more than 5 million messages from over 77,300 users, doubling user reach while cutting support-ticket volume by 71% and live-chat requests by 83% during peak registration periods. Independent local coverage (a sponsored feature on regional broadcaster OFM) confirms the tool's role and origin story but does not itself publish usage statistics.

Caveat: the quantitative results (message volume, user counts, percentage reductions) are self-reported by the commercial vendor that built and sells the platform, and have not been independently audited or confirmed by a third party. They are treated here as descriptive, vendor-reported evidence rather than verified impact data.

Read the full analysis: https://www.ofm.co.za/article/centralsa/332124/kovsiechat-revolutionising-student-engagement

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful