AI Ethics · IBM Framework · POPIA · GDPR

AI Ethics
Framework

Every tool I build and every system I architect is governed by this framework — derived from IBM AI Ethics principles, POPIA legislation, and international best practice.

Established: 19 February 2026 · Aligned with IBM AI Ethics · POPIA Act 4 of 2013 · GDPR 2018

The Five IBM AI Ethics Pillars

My consulting practice and all tools I build are anchored in the five pillars of ethical AI as established by IBM — the organisation that certified my AI expertise.

🔍

Explainability

AI systems I build can explain their reasoning in plain language. No black boxes. Every output can be understood and justified.

IBM AI Ethics · Pillar 1

⚖️

Fairness

AI systems I build are tested for bias and designed to treat all users equitably — regardless of race, gender, age, or geography.

IBM AI Ethics · Pillar 2

🛡️

Robustness

AI systems I build handle unexpected inputs safely, fail gracefully, and maintain reliable performance even in constrained environments.

IBM AI Ethics · Pillar 3

👁️

Transparency

Users always know they are interacting with AI. The source, purpose, and limitations of every AI system I deploy are clearly disclosed.

IBM AI Ethics · Pillar 4

🔒

Privacy

Data minimisation, zero retention, and privacy-by-design are built into every system from the ground up — not added as afterthoughts.

IBM AI Ethics · Pillar 5

My Specific Commitments

Data Sovereignty

Your Data Stays Yours

No data entered into my demo tools is ever stored, shared, or used for any purpose beyond generating your immediate response. Zero retention. By design.

Transparency

Always Disclosed as AI

Every AI tool I build clearly discloses that users are interacting with artificial intelligence. No deception. No impersonation of humans.

Purpose Limitation

Used Only for Stated Purpose

Data collected during consulting engagements is used only for the specific purpose agreed upon in writing. Never repurposed. Never monetised.

Bias Prevention

Actively Anti-Bias

I evaluate AI outputs for cultural, racial, and gender bias before deployment. African contexts are tested specifically — not assumed to match Western defaults.

POPIA Compliance

South African Law First

All tools and consulting services comply with the Protection of Personal Information Act 4 of 2013, including Sections 19 (security) and 22 (notification of breaches).

GDPR Alignment

International Standard

For international clients, all systems are designed to meet GDPR requirements — data minimisation, right to erasure, and explicit consent where required.

I WILL NEVER BUILD AI THAT:

Deceives users into believing they are talking to a human when they are not
Discriminates against individuals based on race, gender, age, religion, or geography
Collects or stores personal data without explicit informed consent
Is designed to surveil, manipulate, or psychologically harm individuals
Serves authoritarian control, mass surveillance, or suppression of rights
Generates or amplifies misinformation, hate speech, or harmful content
Operates without meaningful human oversight in high-stakes decisions

Africa-Specific Ethics Commitments

As an African AI consultant serving African businesses and communities, I hold additional ethical obligations that go beyond global frameworks:

AI systems I build for African clients are tested against African data, languages, and cultural contexts — not just Western benchmarks
I actively work to ensure AI adoption in Africa reduces inequality rather than amplifying existing disadvantages
I price consulting services to be accessible to SMEs and community organisations — not only large corporations
I advocate for African data sovereignty — the right of African communities to control data generated on the continent
I will not build AI systems that exploit African labour, data, or natural resources for offshore benefit without fair compensation

THE MAHLO KGOTLENG ETHICS PROMISE

Every AI system I build, every strategy I recommend, and every tool I deploy will be one I would be proud to demonstrate to any regulator, any community it serves, and any future generation that inherits its impact. I build for dignity. I build for justice. I build for Africa and the world.

Reporting an Ethics Concern

If you believe any tool or service I provide violates these ethics principles, I want to know. Contact me directly at [email protected] with the subject line "Ethics Concern" — I commit to responding within 3 business days and taking appropriate corrective action.

For POPIA-related complaints, you may also contact South Africa's Information Regulator at [email protected] or visit inforegulator.org.za.