How Cam Stevens Thinks About Tech Raising Psychosocial Risk

Episode 15 of Headline Podcast, published on March 18, 2026, brings Cam Stevens together with Andreza Araujo and Dr. Megan Tranter to talk about technology, signal quality and the pressure that builds when tools arrive before the problem is clear. The thesis is simple. Safety tech only earns its place when it reduces risk pressure, changes a real decision and avoids becoming another layer of surveillance.

Cam Stevens says, "If we're very clear on the problem to solve, then a technology catalog is excellent. If we're not clear on the problem to solve, a technology catalog is very dangerous." That sentence is the frame for this companion. The conversation matters because the wrong tool can make leaders feel informed while the underlying work becomes harder to do well.

Why the problem has to come first

The first question is not which platform has the best demo. It is which exposure the organization is trying to reduce, because a tool that does not remove a real pressure point only adds another task for people who already have too many. ISO specifies that ISO 45003:2021 guides psychosocial risk management inside an occupational health and safety system, which is useful here because it forces the discussion back to work design instead of software preference.

Andreza Araujo makes a similar point in A Ilusão da Conformidade, where compliance can look organized while the work itself stays fragile. That is why the episode lands so well for Headline readers. The relevant question is not whether a technology stack is modern. It is whether the stack makes the work safer, easier to govern and less draining on the people who must use it every day.

That logic also sits behind how Cam Stevens thinks about voice technology in EHS, because voice capture is only useful when the signal moves to the right owner and changes what happens next.

Where the psychosocial risk enters

The psychosocial risk appears when a tool changes how people are watched, how fast they must respond and how often they must prove that they are paying attention. Across 25+ years and more than 250 companies in 30+ countries, Andreza Araujo has seen that a tool becomes stressful when it adds friction without adding authority. The worker then gets a new expectation, but not a new way to control the job.

The ILO reports 2.93 million work-related deaths each year, which is enough to keep anyone honest about added complexity. A system that makes people stop, click, log and explain more often than it changes the exposure is not a neutral upgrade. It is a design choice that can widen fatigue, delay action and push the real risk deeper into silence.

That is why this episode belongs beside the 30-day psychosocial risk audit for technology. Leaders need to ask whether the tool helps the crew do the job, or whether it mainly helps the company observe the crew while leaving the work untouched.

Cam Stevens also warns that "the changing shift in risk profile will be overwhelmingly psychosocial, driven by technology usage in our organizations." That is not a prediction for its own sake. It is a warning that monitoring pressure, alert fatigue and reduced autonomy can travel through a plant faster than the intended control does.

What the pilot hides

A pilot can prove that a tool works under attention, but it does not prove that the normal operating system can absorb it. The pilot has vendor support, senior curiosity and a clean launch story. Daily work has night shift, contractor interfaces, weak connectivity, production pressure and supervisors who already carry too many decisions. NIOSH explains Total Worker Health as an approach that connects hazard control with worker well-being, and that helps here because the tool should protect people without becoming a new source of strain.

The danger is that leaders celebrate the pilot before they test the darker conditions. If adoption collapses when the shift is thin, the task is urgent, or the site is under pressure, the pilot has only proven that the tool can work in ideal conditions. It has not proved that it can survive a Tuesday night.

Cam Stevens makes the limit clear when he says, "We can elevate the human experience with technology, but there are certainly times when technology should absolutely not be used." That is the right filter for psychosocial risk because some problems need a human decision, not another interface.

This is also why the voice technology pilot and safety signal governance article belongs in the same reading set. When the signal gets easier to capture, the governance gap gets easier to see.

What good signal governance looks like

Good signal governance starts with ownership, not volume. The organization should know which function acts, how fast it must act and what counts as proof that the field changed. OSHA describes safety and health programs as systems built around management leadership, worker participation, hazard identification and control, which is the right shape for technology governance too. A tool that cannot reach one of those functions is not helping the program. It is only decorating it.

That is where the 24-hour rule matters. If a high-consequence signal arrives and no one can say who must respond before the next shift, the company has a reporting channel without a decision channel. This is the difference between data and governance. The former records pressure. The latter changes it.

It also matters that some tools should not be used in some settings, which is why the episode quote above is more than a sound bite. A system that detects more can still create more fear, more confusion or more second-guessing when leaders have not defined what will happen after the alert. That is where control health versus TRIR versus SIF exposure becomes the better board conversation.

Technology should therefore be treated as a decision aid, not as a substitute for judgment. If a signal does not change supervision, staffing, engineering, escalation or task design, then it is only making the dashboard busier.

Comparison: dashboard tech vs governed tech

The difference between weak and strong use of safety technology is visible in what leaders measure after the system goes live. The weak model counts forms, alerts or downloads. The stronger model counts whether a critical exposure was corrected quickly enough to matter. That is why the conversation around Cam Stevens is useful for EHS managers and safety leaders who need less theater and more proof.

The table matters because technology procurement is usually the easy part. The hard part is deciding what the company will do when the tool reports an inconvenient truth. During the PepsiCo South America tenure, where the accident ratio fell 50% in six months, Andreza Araujo saw that disciplined routines beat slogans. That lesson still applies here. Software does not create discipline. Leaders do.

The same distinction shows up in the voice technology article, where the signal only becomes useful once someone owns the next action. If a board wants the better picture, it should ask what changed in the field, not just how many people clicked the button.

Recommendation

Use Episode 15 as a 30-day audit prompt. Pick one technology already in your stack, define the exposure it should reduce and trace one signal from capture to action. If the path is unclear, the tool is not the problem alone. The governance around the tool is the problem.

Run the review on a normal shift, not only in a clean pilot. Ask who owns the signal, who can stop the job, what changed in the field and how fast the team learned about it. Then check whether the tool reduced pressure or simply moved pressure into a different queue.

That audit is stronger when you compare it with voice technology in EHS and signal governance from the voice pilot. The point is not to buy less technology. The point is to buy only the tools that can change a decision, which is the only standard that keeps psychosocial risk from creeping in through the back door.

Before the next purchase, ask five questions. Which risk does the tool reduce? Which control does it verify? Which person acts? Which decision changes within 24 hours? Which proof shows the field got easier, not just more visible?

Conclusion

Cam Stevens' episode is useful because it does not treat technology as a cure-all. It treats it as a choice that can either relieve pressure or add it, depending on the quality of the problem statement and the strength of the response model.

If your team wants a sharper conversation about safety tech, governance and psychosocial risk, Listen to the full conversation. Episode 15 earns that attention because it shows where the real work begins, which is after the tool is bought and before the team convinces itself that a dashboard is the same thing as control.