Despite the tremendous promise offered by artificial intelligence (AI) for healthcare in South Africa, existing policy frameworks are inadequate for encouraging innovation in this field. Practical, concrete and solution-driven policy recommendations are needed to encourage the creation and use of AI systems. This article considers five distinct problematic issues which call for policy development: (i) outdated legislation; (ii) data and algorithmic bias; (iii) the impact on the healthcare workforce; (iv) the imposition of liability dilemma; and (v) a lack of innovation and development of AI systems for healthcare in South Africa. The adoption of a national policy framework that addresses these issues directly is imperative to ensure the uptake of AI development and deployment for healthcare in a safe, responsible and regulated manner.

How should our patent system deal with inventions by artificial intelligence (AI) systems? The first hurdle to patenting an invention by an AI system is for the AI system to qualify as an inventor in our current patent law. South Africa's Patents Act 57 of 1978 refers to an inventor as 'him'. The Interpretation Act 33 of 1957 provides that reference to the male gender includes the female. However, does 'him' include an AI inventor (as opposed to a human inventor)? If one adopts a literal reading, and assumes that an AI system cannot be referred to as 'him', the answer would be no. But is it the purpose of the Patents Act to only provide patent protection to certain kinds of inventions, namely human inventions, and not AI inventions?This question rapidly entered the realm of reality with the news that South Africa's Patent Office (SAPO), which runs under the auspices of the Companies and Intellectual Property Commission, granted a patent for which the inventor is an AI system. [1][2][3] The AI system, called DABUS (an acronym for 'device for the autonomous bootstrapping of unified sentience'), was created by American AI entrepreneur Dr Stephen Thaler. DABUS invented a new food container, which was the subject of the patent application. 4 In the patent application, Thaler is indicated as the patent owner, and DABUS as the sole inventor. This patent application was not restricted to South Africa. Thaler also submitted the same food container patent application in various other jurisdictions. In light of the fact that DABUS's food container patent application had already been rejected by the leading patent offices of the world, namely the European Patent Office (EPO) 5 and the United States Patent and Trademark Office (USPTO) 6 , SAPO's decision to grant a patent, for which DABUS is the inventor, was received with a mixture of fascination and disbelief by the intellectual property (IP) community. 1,7,8 In fact, some commentators even thought that the SAPO decision was an error, or an oversight due to South Africa's formal (but not substantive) examination system for patent applications. [7][8][9] In this Commentary, we suggest that the SAPO decision -whether intentional or not -was the right decision from a legal perspective.

This paper reports the results of a public engagement study on heritable human genome editing (HHGE) carried out in South Africa, which was conducted in accordance with a study protocol that was published in this journal in 2021. This study is novel as it is the first public engagement study on HHGE in Africa. It used a deliberative public engagement (DPE) methodology, entailing inter alia that measures were put in place to ensure that potential participants became informed about HHGE, and that deliberations between the participants were facilitated with the aim of seeking consensus. A diverse group of 30 persons was selected to participate in the DPE study, which took place via Zoom over three consecutive weekday evenings. The main results are: Provided that HHGE is safe and effective, an overwhelming majority of participants supported allowing the use of HHGE to prevent genetic health conditions and for immunity against TB and HIV/Aids, while significant majorities opposed allowing HHGE for enhancement. The dominant paradigm during the deliberations was balancing health benefits (and associated improvements in quality of life) with unforeseen health risks (such as loss of natural immunity). The seriousness of a health condition emerged as the determining factor for the policy choice of whether to allow an application of HHGE. More generally, equal access to HHGE qua healthcare service featured as an important value, and it was uncontested that the South African government should allocate resources to promote scientific research into HHGE. These results are aligned with the policy principles for regulating HHGE in South Africa suggested by Thaldar et al. They call for urgent revision of South African ethics guidelines that currently prohibit research on HHGE, and for dedicated HHGE legal regulations that provide a clear and comprehensive legal pathway for researchers who intend to conduct HHGE research and clinical trials.

Questions have arisen as to the agent status of artificial intelligence (AI) systems. This article endeavors to build a general framework for reframing what “agent” and “agency” means to a more inclusive approach. This article further demonstrates the issues in current approaches (like the mental or intentional approaches) serving to justify the proposed change. It begins with the theoretical underpinnings of ‘explanations’ and how these functions to build theories. Thereafter a model of agency based on the notion of “purposive action” is built. Purposive action is understood to be agents who utilize features in their environments as affordances which are conducive towards their goals. Furthermore, this article builds a notion of agency and autonomy based on semantic meaning which is an intrinsic feature of complex systems themselves. This framework, along with the methods presented provides for a novel framing of agency and a means to rank agents and agency status.

No abstract