Amalta Seevnarayan Dhanee | Legal Advisor

In an era where artificial intelligence (AI) is rapidly transforming the landscape of communication, ChatGPT an AI Bot, launched in 2020, emerges as a powerful and versatile tool.

From answering queries to composing poetry, this AI language model has become an integral part of our digital interactions.

However, as its influence grows, so do the legal questions surrounding its use. In South Africa, a nation at the crossroads of technological advancement and legal tradition, the arrival of ChatGPT raises complex issues that demand thoughtful consideration.

In this article, we delve into the legal ramifications of ChatGPT, exploring topics such as liability, privacy, and intellectual property within the South African context.

Whilst this AI Bot is a game changer for many around the globe, it relies solely on human interaction to feed it information and learn new facts. It lacks human qualities such as emotional intelligence, critical thinking and creativity, as a result one must exercise caution when using ChatGPT and consider the legal ramifications thereof.

Legal Ramifications to Consider When Utilising ChatGPT

1. Breaching Provisions of Protection of Personal Information Act 4 of 2013 (“the POPI Act”)

The purpose of the POPI Act is to, inter alia, give effect to the constitutional right to privacy, by safeguarding the processing of all personal information in such a way that it balances the right to privacy against other rights (such as the right of access to information), and protects important interests, including the free flow of information within South Africa and across international borders.

Should a user fail to adhere to the provision of the POPI Act on the ChatGPT platform, he/she will be held accountable in terms of the Act. This can include an administrative fine not exceeding
R10 000 000.00.

2. Ability to generate inaccurate information/fake news

There have been a number of cases in which ChatGPT proved to be unreliable. In many instances it referred to case law which did not exist and provided incorrect case citations.

In another instance it referred to an article published in the Washington Post, when in fact no such article had ever been published. In response to the fabricated article, the spokesperson for ChatGPT stated that “When users sign up for ChatGPT, we strive to be as transparent as possible that it may not always generate accurate answers. Improving factual accuracy is a significant focus for us, and we are making progress.”

Placing reliance on information obtained via ChatGPT without verifying the accuracy thereof can lead to reputational damage.

3. Legal documents and Agreements

Drafting of legal documents and agreements is a popular feature of ChatGPT and has made the ordinary man feel like an expert in drafting these documents and agreements, thereby eliminating the need to consult with an attorney.

The danger herein lies in that the ordinary man is not au fait with the relevant legislation and/or laws affecting his agreement/document and will accept the agreement/document generated as being accurate. Furthermore, material terms and conditions may be omitted. This could result in an agreement that is not enforceable or legally sound.

The safest option would be for the user to engage the services of an attorney to peruse the agreement, ensuring it is legally sound and enforceable.

4. Infringement of Intellectual Property Rights

As previously stated, ChatGPT and other AI Bots rely on human interaction to feed it information. If it is fed with documents/materials/books that are protected by copyright laws, it could very well infringe on the copyright laws when responding to a user. The user, in turn, could possibly find himself in hot water for contributing to the infringement.  

The advent of ChatGPT introduces both promise and peril.

While it streamlines research, augments drafting, and enhances efficiency, it also raises ethical and professional challenges. Legal professionals must grapple with questions of accountability, transparency, and the delicate balance between human judgment and machine-generated insights.

As ChatGPT becomes an indispensable tool, it is incumbent upon lawyers, judges, and policymakers to navigate this uncharted territory. How do we attribute legal advice when it emanates from an algorithm? Can ChatGPT be held liable for errors in legal analysis? What safeguards can we implement to protect client confidentiality?

Legal practitioners should embrace ChatGPT as a powerful ally, but with discernment. We must wield it ethically, ensuring that justice remains paramount. As we forge ahead, let us remember that the legal profession is not merely about statutes and precedents; it is about empathy, wisdom, and the pursuit of a just society.