2025 in Review

The EHS3 Shifts That Quietly Changed How Work Gets Done

Most years, progress in Environment, Health, Safety, Sustainability, and Security shows up as new tools, new rules, or new language.

2025 was different.

What changed this year was not ambition. It was honesty.

Across North America, organizations stopped pretending that safety, sustainability, and security live in binders or dashboards. They started treating EHS3 as an operational condition that either supports work or quietly undermines it.

I have yet to walk into a facility that feels unsafe because people do not care. What I keep seeing instead are systems that lag behind reality. 2025 exposed that gap more clearly than any year I can remember.

Here are the ten EHS3 shifts from 2025 that actually mattered and what operations leaders can start doing with them now.

1. AI-Driven EHS3 Management Systems

Prediction beats perfection

Artificial intelligence entered EHS3 systems in a real way this year. Not as flashy dashboards, but as tools that surface weak signals. Patterns that humans miss. Trends that show up weeks before an incident, not weeks after.

Here is the uncomfortable truth. Most organizations already have enough data to prevent their next serious incident. They just do not look at it together.

The value of AI is not intelligence. It is integration.

What I see on the floor

Near misses logged with no follow-up. Maintenance delays noted but not escalated. Audit findings closed on paper while the condition remains.

AI only works if leaders are willing to act on what the system reveals.

What to do now

Clean your data before buying tools. Require clear narratives in incident and near-miss reporting. Ask one simple question every month: what risk keeps showing up in different places under different names?

2. Wearables and IoT Monitoring

Visibility changes behavior

Wearables became common in higher-risk environments this year. Not because they are perfect, but because they reveal how work actually happens when supervisors are not standing nearby.

What surprised many leaders was not the number of alerts. It was when and where work drifted from plan.

I have seen wearables confirm what workers have been saying for years about fatigue, heat, and task stacking.

The mistake to avoid

Using data as discipline.

The moment workers think monitoring is about catching mistakes, the value disappears.

What to do now

Pilot wearables in one task where risk is already accepted. Be explicit about how data will and will not be used. Treat findings as design feedback, not performance feedback.

3. Immersive and Remote Training

Training finally started matching reality

VR and scenario-based training took off because they address a long-standing problem. Most training teaches rules. Real work requires decisions.

When training mirrors the pressure and ambiguity of the job, retention follows.

What I keep seeing

People pass training and still improvise because training never showed them what failure looks like.

What to do now

Identify tasks where failure is rare but unforgiving. Run tabletop scenarios even without VR. Ask operators what part of the job makes them hesitate. Train there.

4. Mental Health as an EHS3 Factor

Fatigue is not a personal issue

2025 made it harder to ignore what many supervisors already knew. Stress, fatigue, and distraction precede incidents.

We like to call them human factors. In reality, they are system outputs.

What I see repeatedly

Long hours normalized. Breaks treated as optional. Schedule changes made without considering cognitive load.

Then we act surprised when someone makes a mistake.

What to do now

Watch overtime trends the same way you watch injury rates. Give supervisors language to pause work without escalation. Treat mental strain as a leading indicator, not a soft topic.

5. ESG Integrated into EHS3

This stopped being corporate language

In 2025, ESG stopped living in slide decks and started touching daily work. When emissions, waste, and labor conditions are managed together, tradeoffs become visible.

Most plants already influence ESG outcomes whether they track them or not.

What goes wrong

Goals set at the top without understanding process reality.

What to do now

Map where safety and sustainability intersect. Involve operators in setting reduction targets. Measure fewer things better.

6. Net-Zero Commitments

Operations owns the outcome

Net-zero commitments expanded rapidly this year. What became obvious is that these goals succeed or fail in maintenance shops, process changes, and capital planning meetings.

This is not a communications effort. It is operational discipline.

What to do now

Start with energy visibility. Tie efficiency projects to reliability. Involve maintenance before committing to timelines.

7. Renewable Energy and Storage

Resilience became the real driver

Renewables crossed major thresholds in 2025, supported by energy storage growth. What mattered operationally was not ideology. It was stability.

Facilities with diversified energy options had fewer surprises.

What to do now

Understand your peak demand exposure. Explore storage as a risk-reduction tool. Include energy in capital planning conversations early.

8. Strengthened Safety Regulations

Rules finally caught up

Updates to PPE fit, hazard communication, and heat illness reflected what many sites already struggled with.

Compliance got more practical.

What I see often

Rules applied evenly to uneven work.

What to do now

Re-evaluate PPE fit across all body types. Use real chemicals from your site in training. Plan work around heat instead of reacting to it.

9. Plastics and Waste Reduction

Waste is a design signal

Waste reduction efforts in 2025 exposed inefficiencies upstream. Disposables usually exist because something else is hard.

What to do now

Ask why disposables are used, not just how many. Bring procurement and operations together. Treat waste as lost efficiency, not just environmental impact.

10. Cybersecurity as an EHS3 Risk

Digital failures can cause physical harm

Cybersecurity became an EHS3 issue because operational systems are now safety systems.

I have yet to meet an operations leader who believes a cyber event could not affect their site. Most just do not know where the exposure lives.

What to do now

Inventory control systems. Include cyber scenarios in emergency planning. Hold vendors accountable for secure design.

The Throughline

Integration beats innovation every time

What stands out about 2025 is not the technology. It is the shift toward reality.

The organizations making progress are not chasing trends. They are fixing interfaces. Between systems. Between departments. Between intent and execution.

Safety improves when work is designed to fail safely. Sustainability improves when waste is treated as feedback. Security improves when digital systems are treated like physical ones.

Here is the question worth sitting with:

Which EHS3 risks in your operation are quietly tolerated because “that’s just how work gets done”?

That answer usually tells you exactly where to start.


Originally published January 3, 2026 in The All-In on Safety Brief.

Explore the SoIN Safety approach · Request a Diagnostic Conversation

Previous
Previous

No Action Is Complete Without Proof.

Next
Next

The First Lesson Every Safety Leader Should Learn When Joining a New Company