20/07/2026
AI is officially a workplace safety issue now, not just an IT one.
Since February, NSW law has named AI and algorithms as things employers have to actively manage risk for. If a rostering algorithm or an automated system puts a worker at risk, that responsibility sits with the business, not the software. Although this is only in NSW, watch this space for other states and territories, as it's only a matter of time
Here's the thing though: AI cuts both ways on safety. Used well, it's catching things humans miss. Cameras flagging missing PPE the second someone steps into a hazard zone. Vision systems picking up structural corrosion or equipment wear between scheduled inspections. Rostering tools flagging fatigue risk before it becomes an incident. That's real, measurable safety improvement.
But the same automation can create risk nobody planned for. A roster optimised purely for coverage can be technically fine and still exhausting. Constant monitoring changes how people feel at work, even when it's there for good reason. That's a psychosocial hazard, same as any other, and it needs managing the same way.
The question for any business running these tools: do you actually know what your digital systems are doing to your people? Worth a proper look before a regulator asks you first.
Get in touch if you want to talk through what this means for your site. 👷♂️