South African lawyers have cautioned employees that using AI tools such as ChatGPT, Gemini, Claude, Grok and Google Gemini at work may create legal exposure for their employers.

Bowmans, a South African law firm, highlighted that AI‑driven recruitment and HR decisions could breach the Protection of Personal Information Act (POPIA).

Under POPIA Section 71, decisions with significant effects on individuals cannot be based solely on automated processing, meaning a human must be able to review and challenge AI‑generated outcomes.

The firm warned that uploading personal or confidential information to public AI platforms could make the employer liable, as the employer determines the purpose and means of data processing.

Bowmans noted specific risks when AI tools are used to screen CVs, conduct chatbot interviews or generate performance ratings that could affect promotions or dismissals.

The lawyers advised companies to set clear AI usage policies, limit the type of information staff can share with such tools, and train employees on their obligations under POPIA.

South Africa currently has no dedicated AI legislation, so existing privacy and employment laws apply to AI‑related workplace practices.