Regulation Of Future Autonomous Mobility Law in SOUTH AFRICA

Regulation of Future Autonomous Mobility Law in South Africa

Introduction

Autonomous mobility refers to transportation systems that operate with minimal or no human intervention through artificial intelligence (AI), machine learning, sensors, robotics, vehicle-to-vehicle (V2V), and vehicle-to-infrastructure (V2I) technologies. These include:

  • Autonomous vehicles (AVs)
  • Connected and automated vehicles (CAVs)
  • Autonomous public transport
  • Self-driving freight systems
  • Autonomous drones and delivery vehicles

South Africa presently has no comprehensive legislation specifically regulating fully autonomous vehicles, although existing transport, consumer protection, constitutional, delictual, insurance, and data protection laws would apply. Legal scholars and policymakers increasingly recognize the need for a dedicated regulatory framework for autonomous mobility.

1. Current South African Legal Framework

Future autonomous mobility will likely be governed through a combination of existing legislation:

A. Constitution of the Republic of South Africa, 1996

Relevant constitutional rights include:

  • Right to life (Section 11)
  • Human dignity (Section 10)
  • Privacy (Section 14)
  • Equality (Section 9)
  • Access to courts (Section 34)
  • Just administrative action (Section 33)

Autonomous systems making transport decisions must comply with these constitutional guarantees.

B. National Road Traffic Act 93 of 1996

The Act currently assumes:

  • Presence of a human driver
  • Human control of a vehicle
  • Human fault in road incidents

Autonomous vehicles challenge these assumptions because control may shift from the driver to software systems. Legal commentators note that South Africa lacks explicit provisions governing autonomous vehicles under current road traffic legislation.

C. Consumer Protection Act 68 of 2008

Manufacturers may incur liability for:

  • Defective autonomous software
  • Sensor failures
  • Algorithmic malfunctions
  • Cybersecurity vulnerabilities

The CPA introduces strict product liability principles that could apply to autonomous vehicles.

D. Protection of Personal Information Act (POPIA)

Autonomous mobility systems process:

  • Location data
  • Biometric data
  • Driving behaviour
  • Facial recognition information

POPIA requires lawful processing, transparency, and accountability for such data. Emerging South African AI governance discussions emphasize accountability for automated decision-making.

2. Key Regulatory Challenges

A. Determining the "Driver"

A fundamental question is:

Who is the legal driver?

Potential answers include:

  • Vehicle owner
  • Passenger
  • Manufacturer
  • Software developer
  • AI system operator

Current South African law recognizes only natural or juristic persons, not autonomous software systems.

B. Accident Liability

Potentially liable parties may include:

  • Vehicle manufacturers
  • Software developers
  • Fleet operators
  • Infrastructure providers
  • Owners

Future legislation will need clear allocation of fault among these actors.

C. Cybersecurity

Autonomous mobility depends heavily on software and connectivity.

Risks include:

  • Hacking
  • GPS spoofing
  • Sensor manipulation
  • Data theft

Future legislation is likely to impose cybersecurity obligations similar to international standards governing automated driving systems.

D. Algorithmic Decision-Making

Questions arise regarding:

  • Ethical decision-making
  • Bias in AI systems
  • Explainability of vehicle decisions
  • Transparency requirements

These concerns overlap with broader AI governance developments in South Africa.

3. Future Regulatory Model for South Africa

Legal scholars suggest that South Africa will likely adopt a framework based on:

Licensing and Certification

Before deployment, autonomous vehicles may require:

  • Safety certification
  • Software validation
  • Cybersecurity assessment
  • Regulatory approval

Data Recording Requirements

Autonomous vehicles will likely require:

  • Event data recorders
  • Black-box systems
  • Incident logs
  • Algorithmic audit trails

Mandatory Insurance

Future legislation may require:

  • Manufacturer insurance
  • Operator insurance
  • Autonomous mobility compensation funds

Regulatory Oversight

Potential regulators include:

  • Department of Transport
  • South African Bureau of Standards
  • Information Regulator
  • South African Police Service
  • Competition authorities

4. Important South African Case Law Influencing Future Autonomous Mobility Regulation

Although South Africa has few cases specifically concerning autonomous vehicles, several landmark judgments establish principles that will govern future autonomous mobility.

Case 1: Minister of Safety and Security v Van Duivenboden

Principle

The State may be liable where it fails to act reasonably to prevent foreseeable harm.

Relevance to Autonomous Mobility

If regulators authorize unsafe autonomous technologies without adequate safeguards, governmental liability may arise for foreseeable accidents.

Case 2: Carmichele v Minister of Safety and Security

Principle

Constitutional values influence the development of common law duties.

Relevance

Courts may develop new duties concerning:

  • AI safety
  • Automated transport systems
  • Public protection

where existing law proves inadequate.

Case 3: Loureiro v Imvula Quality Protection

Principle

Negligence arises where reasonable precautions against foreseeable harm are not taken.

Relevance

Manufacturers of autonomous vehicles may be liable if:

  • Software defects were foreseeable
  • Cybersecurity weaknesses were ignored
  • Safety testing was inadequate

Case 4: MEC for Education, KwaZulu-Natal v Pillay

Principle

Decision-making systems must respect constitutional rights and fairness.

Relevance

Autonomous transport systems making automated decisions must comply with constitutional principles of fairness and equality.

Case 5: Le Roux v Dey

Principle

Technological conduct causing reputational harm can attract legal liability.

Relevance

Autonomous mobility systems collecting, sharing, or processing personal information could create liability where privacy or dignity rights are infringed.

Case 6: K v Minister of Safety and Security

Principle

Institutions may bear vicarious liability for harmful conduct connected to their functions.

Relevance

Fleet operators and mobility providers may be liable for harm caused by autonomous systems operating within their control structures.

Case 7: Minister of Finance v Gore

Principle

Pure economic loss caused through negligent conduct may be recoverable.

Relevance

Businesses suffering losses from autonomous system failures may seek damages against operators or manufacturers.

Case 8: Barkhuizen v Napier

Principle

Contractual terms must comply with constitutional values and public policy.

Relevance

Autonomous vehicle manufacturers may not be able to exclude liability through unfair contractual clauses.

5. Future Liability Model

South Africa is likely to adopt a mixed liability system involving:

ActorPossible Liability
ManufacturerProduct defects
Software developerCoding failures
Fleet operatorOperational negligence
Vehicle ownerMaintenance failures
Infrastructure providerConnectivity defects
Government regulatorRegulatory negligence

6. International Influence on South African Regulation

South Africa is expected to draw from:

  • UNECE Automated Driving Regulations
  • European Union AI and vehicle safety frameworks
  • German Autonomous Vehicle Act
  • United States autonomous vehicle testing regulations

Research on autonomous vehicle regulation in South Africa consistently recommends adapting foreign models while addressing local infrastructure, safety, and governance realities.

Conclusion

The future regulation of autonomous mobility in South Africa will require a comprehensive legal framework addressing:

  1. Autonomous vehicle licensing and certification.
  2. Product and software liability.
  3. Cybersecurity obligations.
  4. Data protection and privacy.
  5. AI accountability and transparency.
  6. Insurance and compensation mechanisms.
  7. Constitutional safeguards for safety, dignity, equality, and privacy.

While South Africa currently lacks dedicated autonomous vehicle legislation, existing constitutional, delictual, consumer protection, and administrative law principles—illustrated through cases such as Carmichele, Van Duivenboden, Loureiro, K v Minister of Safety and Security, Le Roux v Dey, and Barkhuizen—provide the foundational legal framework upon which future autonomous mobility regulation is likely to be built.

LEAVE A COMMENT