Skip to main content

Tsholofelo Setati AI Strategy & Innovation Economics

About
Expertise
Economics
Tech + Policy
Insights
Demos
Contact

AFRICA SUBSEA

180 Tbps

+21% YoY as of 2026-01-15

AU AI STRATEGY

Day 650

since adoption 2024-07-19

EU AI ACT D-DAY

T-94 days

until high-risk obligations apply

FRONTIER ¢/1M TOK

—

computing… frontier tier

… ····· … ····· … ····· … ····· … ····· … ·····

TS

About
Expertise
Economics
Tech + Policy
Insights
Demos
Contact

Diagnostic

AI Maturity Diagnostic

Ten questions across six dimensions. The result is a profile and three priority recommendations. Everything runs in your browser. No data leaves the page.

0 / 10 answered

Q1 AI Strategy

How clearly defined is your enterprise AI ambition?

No formal ambition exists A few teams have AI plans, no enterprise view A documented AI strategy exists but is not actively used A board-endorsed strategy guides funding decisions Strategy is reviewed quarterly and tied to enterprise OKRs

Q2 AI Strategy

Who owns AI outcomes at the executive level?

Nobody, AI is a side initiative IT leadership, by default A nominated AI lead reporting to the CIO A dedicated executive (Chief AI Officer or equivalent) Distributed across the C-suite with shared accountability

Q3 Governance & Responsible AI

How are AI use cases assessed for risk before deployment?

They are not Ad hoc legal review A risk checklist exists but is rarely enforced A risk gate is part of the delivery process Tiered risk model with mandatory evaluation harnesses

Q4 Governance & Responsible AI

How do you evaluate model and system quality in production?

We do not Spot checks by the delivery team Manual review on a defined cadence Automated evaluation suites for each release Continuous evaluation tied to alerting and rollback

Q5 Platform & Architecture

Do teams share a common AI platform?

Every team builds from scratch Some shared libraries, no platform team Shared LLM gateway and observability Opinionated platform with golden paths Platform is product-managed with SLOs and a roadmap

Q6 Data Foundations

How accessible is high-quality data to AI delivery teams?

Major friction; data is siloed Possible but slow; bespoke pipelines per use case A central data catalogue exists Self-service access through governed products Productised, contract-backed data with lineage

Q7 Talent & Operating Model

How do you staff AI delivery?

Whoever is available A small central team that takes on every use case Embedded squads with shared platform support Squads paired with business owners and SREs Rotating staffing model with deliberate capability transfer

Q8 Value Realisation

How are AI investment cases evaluated?

Strategic value statements only Single-point ROI estimates Scenario-based business cases Risk-adjusted IRR with explicit assumptions IRR + kill-switch + 6/12-month re-baselining

Q9 Value Realisation

Are run costs (inference, observability, retraining) modelled?

No Estimated as a % of build cost Modelled per use case Modelled with sensitivity analysis Modelled, monitored against actuals, FinOps embedded

Q10 Platform & Architecture

What is the typical idea-to-production cycle time for an approved use case?

Over 12 months 6–12 months 3–6 months 6–12 weeks Under 6 weeks

Your AI maturity profile

0% …

Overall maturity

Dimension breakdown

Priority recommendations

Want a deeper read of these results, with a tailored roadmap?

Book a discovery conversation

AI. Technology Economics. Governance. Strategy.

Johannesburg, South Africa

Site

About
Experience
Expertise
Contact

Work

Insights
Tech + Policy
Economics
Demos
AI Diagnostic
How AI is used here

Connect

LinkedIn
Email
RSS

Stay in the loop

A newsletter on AI strategy, innovation economics, and what it means for African enterprise. Coming soon.

© 2026 Tsholofelo K. Setati. All rights reserved. Built with Astro · How AI is used here

Esc

↑↓ navigate Enter open Esc close

Welcome

Tsholofelo K. Setati

AI. Technology Economics. Governance. Strategy.

Translating emerging technology into compounding business advantage - with field notes, working demos and a clear point of view.

Who I am Background and how I work Working demos Prototypes you can open in the browser Field notes Long-form thinking on AI and innovation

Press ⌘K any time to search

TK

TsholofeloK.Setati

AI. Technology Economics. Governance. Strategy.