AI governance isn’t just a conversation for technology companies. For manufacturing, construction, infrastructure and utilities, the stakes are often much higher. When AI gets it wrong in these environments, the consequence can be a safety incident, operational disruption, or an asset failure.

This is where I believe we need to shift the conversation. 

Much of the guidance around AI governance assumes organisations are building AI models. Most operational businesses aren’t. They’re deploying AI into existing environments where decisions have real-world consequences. 

Think about where AI is already showing up: 

• Computer vision monitoring site PPE compliance and hazard zones. 
• Predictive maintenance influencing critical equipment servicing. 
• Autonomous and semi-autonomous plant and machinery. 
• AI-assisted scheduling across utilities and logistics networks. 
• When these systems fail, the impact becomes operational. 

The future belongs to organisations that are future ready—those that recognise new risks early and build them into the governance systems they already trust.

 

For critical infrastructure operators, this conversation is no longer theoretical. Australia’s Enhanced Critical Infrastructure Risk Management Program Rules, introduced under the SOCI Act in June 2026, now require operators across energy, electricity, gas, water and freight to identify and manage risks associated with novel and emerging technologies, including AI. Compliance begins from 2027. In the United States, NIST is moving in the same direction with its proposed AI Risk Management Framework profile for critical infrastructure. 

The message is remarkably consistent. AI isn’t creating a separate category of governance. Existing risk frameworks already apply, and leadership remains accountable. 

For operational organisations, that means your strongest foundation is likely the one you already have. 

• Your work health and safety systems. 
• Your asset integrity processes. 
• Your operational risk framework. 
• Your lines of accountability. 

The opportunity isn’t to build another framework. It’s to make your existing one Future Fit. 

Here are three questions every operational leader should be asking today: 

1. Where is AI already influencing safety, maintenance or operational decisions? 
2. Who owns the risk when an AI-assisted operational decision is wrong? 
3. Does your enterprise risk framework explicitly recognise AI as an emerging operational risk? 

Leadership in uncertain times isn’t about reacting once regulation arrives or after an incident occurs. It’s about putting risk into strategy before someone else forces the conversation. 

The organisations that thrive won’t be those with the biggest AI investment. They’ll be the ones that build governance that is ready for whatever the future may bring.