5.8 — Skills for building and operating AI

AI redistributes responsibilities among people who understand the business, data, software, safety and compliance.

Four role families

• Building — AI engineers assemble models, data, tools, interfaces and evaluations.
• Operations — AI platform and MLOps teams deploy, monitor, version and restore systems.
• Governance — AI, risk, legal and privacy owners define authorised uses, evidence and controls.
• Security and quality — security, red-team and evaluation roles find weaknesses and track incidents.

In a small organisation, one person may cover several responsibilities, but none should disappear.

Read numbers without distorting them

The World Economic Forum’s Future of Jobs Report 2025 contains employer-survey forecasts, not a guaranteed count of future jobs. PwC observed a 56% average wage premium in 2024 for jobs requiring AI skills. This is a point-in-time comparison, not annual salary growth or an individual promise.

Sources: WEF — Future of Jobs Report 2025 and PwC — Global AI Jobs Barometer 2025.

✅ Summary

• AI systems require building, operations, governance, security and quality responsibilities.
• Present every study with its population, date and whether it is an observation or forecast.
• PwC’s 56% is a wage premium comparison, not salary growth.