How Technology Is Quietly Reshaping Workplace Harm

For years, we’ve talked about technology as if it’s a magic fix for safety: sensors, automation, AI, wearables, digital reporting. And yes – these tools can absolutely reduce harm.

New technology doesn’t just solve old risks – it creates new ones. And most organisations aren’t ready for them. We’re entering a world where harm won’t always look like a crushed hand or a fall from height. It will look like privacy breaches, digital manipulation, cognitive overload, algorithmic bias, and new forms of exploitation.

Let’s talk about the risks that are coming – and the ones already here.

Surveillance Tech & AI Wearables: When “Safety” Becomes a Cover for Control

Take the example you raised: AI glasses in childcare centres.

On paper, they’re sold as:

• improving documentation
• reducing admin
• capturing real‑time learning moments

But from an H&S perspective, they also introduce:

• privacy risks
• exploitation risks
• misuse of footage
• worker stress from constant monitoring
• children being recorded without meaningful consent

The risk isn’t the device – it’s the power imbalance it creates.

And this won’t be limited to childcare. Expect similar issues in:

• aged care
• hospitality
• retail
• logistics
• construction
• manufacturing

Anywhere a worker can be monitored, someone will try to monitor them.

Deepfakes, Voice Cloning & Digital Manipulation

This is a risk category almost no H&S system is prepared for.

Emerging threats include:

• fake instructions given to workers via cloned supervisor voices
• deepfake training videos spreading incorrect procedures
• fraudulent authorisations for hazardous work
• reputational harm to workers through manipulated footage

Imagine a worker receiving a voice message that sounds exactly like their manager telling them to override a lockout. That’s not science fiction – it’s already possible.

This becomes a critical risk when it intersects with:

• confined spaces
• energy isolation
• hazardous substances
• plant and machinery

H&S systems will need authentication protocols, not just procedures.

Cognitive Overload & Decision Fatigue From Too Much Tech

We’ve spent years telling workers to “stay alert”, “follow the procedure”, “check the app”, “log the hazard”, “scan the QR code”, “watch the dashboard”.

Now add:

• wearables
• alerts
• notifications
• digital forms
• AI prompts
• automated warnings

The risk? Cognitive overload becomes a safety hazard in itself. Workers can’t meaningfully process 200 alerts a day. Supervisors can’t interpret 10 dashboards at once. Leaders can’t make good decisions drowning in data.

The future risk isn’t lack of information – it’s too much of it.

Algorithmic Bias in Safety‑Critical Decisions

AI systems are already being used to:

• prioritise risks
• flag “unsafe behaviour”
• analyse incident patterns
• identify “high‑risk workers”
• recommend controls

But algorithms reflect the data they’re trained on.

Which means:

• bias gets baked into safety decisions
• certain groups may be unfairly labelled “risky”
• disciplinary action may be triggered by flawed data
• controls may be misallocated
• psychosocial harm increases

This is a new category of harm: algorithmic discrimination. And it sits squarely within HSWA’s definition of health.

Automation Blindness: When People Trust Tech Too Much

As automation increases, so does a dangerous mindset: “The system will catch it.” We’ve already seen this in aviation, transport, and healthcare.

Emerging risks include:

• workers assuming sensors will detect hazards
• operators relying on automated shutdowns
• maintenance teams trusting digital diagnostics over physical checks
• supervisors believing dashboards instead of workers

Automation reduces some risks but creates new failure modes:

• cascading system errors
• silent faults
• software bugs
• sensor drift
• cyber interference

When humans stop noticing because they assume the machine will – that’s a critical risk.

Cybersecurity as a Physical Safety Issue

A hacked system isn’t just an IT problem.

It can:

• disable safety interlocks
• override machine controls
• shut down ventilation
• manipulate chemical dosing
• alter traffic management systems
• lock workers inside access‑controlled areas

Cyber risk is now a physical risk. H&S and IT can’t operate in separate universes anymore.

New Forms of Psychosocial Harm

Technology is accelerating psychosocial risk in ways we haven’t fully named yet.

Emerging exposures include:

• AI performance scoring
• constant digital monitoring
• fear of being replaced by automation
• pressure to respond instantly
• loss of autonomy
• digital harassment
• online abuse from customers
• AI‑generated misinformation about workers

These aren’t “HR issues”. They’re health issues.

The Blurring of Work & Life Through Tech

Remote work, gig platforms, and digital tools have dissolved boundaries.

New risks include:

• fatigue from always‑on communication
• unsafe home “work” environments
• isolation and reduced peer support
• blurred responsibility for equipment and ergonomics
• increased exposure to domestic violence risks
• gig workers falling outside traditional H&S protections

Technology has changed where work happens – but legislation and systems haven’t caught up.

So What Does the H&S Profession Need to Do?

Expand the definition of “harm”

Include digital, psychological, ethical, and privacy‑related risks.

Build capability in tech literacy

H&S professionals must understand:

• AI
• automation
• data ethics
• cybersecurity
• digital systems

Strengthen worker voice

Technology must be implemented with workers, not to them.

Develop new controls for new risks

Not just PPE and procedures – but:

• authentication protocols
• data governance
• ethical frameworks
• transparency requirements
• digital fatigue management

Challenge tech vendors

Ask:

• What harm could this create?
• How is data used?
• Who has access?
• What happens if it fails?

Keep the human at the centre

Technology should enhance safety, not erode dignity.

Changing Landscape

The next decade of workplace harm won’t be defined by machinery, heights, or manual handling.

It will be defined by:

• data
• algorithms
• surveillance
• automation
• digital manipulation
• psychological pressure
• ethical blind spots

The organisations that thrive will be the ones that recognise this early – and evolve their H&S systems before the risks evolve faster than they can respond.