Why Does Staff Wellbeing Miss Online Harassment at Work?

Short answer: Harassment at work used to happen out loud, in a room, in front of people who could later say what they saw. It now happens mostly on devices, in group chats nobody at work administers, through images made by an app, and increasingly through cameras worn on somebody’s face. Most staff wellbeing plans name none of this, so the people affected assume the organisation would rather not hear about it.

In short:

• Harassment has moved onto personal phones and private chats, where no manager can see it and nothing is kept.
• Anyone can now build a convincing explicit image of a named colleague from an ordinary photograph on your website.
• Glasses with built-in cameras look like ordinary glasses, so nobody in your building can tell who is recording whom.
• HR can handle the process side of all this well, but cannot be expected to know how the technology works.
• Naming these behaviours out loud, in training, is what makes people start reporting them.

How harassment moved off the office floor

Twenty years ago, workplace harassment happened in front of witnesses. Somebody said something in a corridor or a meeting, other people heard it, and an investigation could ask them what they remembered. The behaviour was ugly, but at least it was visible.

That has changed almost completely, and it changed quietly. Most working conversation now runs through messaging apps, and a large share of it runs through groups that the organisation never set up. A team leader creates a WhatsApp group on a personal phone, everybody joins, and the working day carries on inside it. Nobody administers that group, nobody retains anything from it, and no manager will ever see what happens there unless a member chooses to forward something out.

Because the venue changed, the behaviour changed with it. Harassment on a screen is far more deniable than harassment in a corridor. Somebody frames a comment as a joke. A video arrives with “somebody just sent it over”, as though that settles the question of who chose to share it. Two colleagues put a thumbs-down on every message one person posts, which everyone in the chat understands perfectly well and nobody can quote in a complaint. Messages arrive at eleven at night, reaching the person at home, on the phone they use for everything else in their life.

None of that is less serious than a comment in a corridor. It is simply much harder for you to see, and much harder for the person on the receiving end to prove.

Where deepfakes fit into this

The same private chats that carry the comments now carry something newer, and considerably worse.

A deepfake is a fake image or video of a real person, produced by an app. In a workplace it almost always means an explicit image of a named colleague, and the important thing to understand is how ordinary the raw material is. Someone takes a headshot from your website, a photograph from a company LinkedIn post, or a picture from the summer party. They feed it into an app and get back something realistic enough to circulate. It takes a few minutes and no technical skill whatsoever.

From there it travels exactly like everything else, through the group chats described above, and the person it depicts is usually the last to learn it exists. That combination, easy to make and easy to spread, is why this has moved so quickly from a curiosity to something employers now have to think about.

The rules have started to catch up. The Government says it will make creating sexually explicit deepfake images a criminal offence, and notes that women and girls are usually the target. On workplace harassment more broadly, Parliament passed the Worker Protection (Amendment of Equality Act 2010) Act 2023, and ACAS sets out in plain terms what it expects employers to do. Your own advisers will tell you what any of that means for your organisation, because those are legal questions rather than technical ones.

The technical question is the one that tends to go unanswered. What does this look like when it lands, how far has it already travelled, and what should somebody do in the first hour? Harassment training written even three years ago says nothing at all about it.

Cameras that do not look like cameras

If a deepfake starts with a photograph, the obvious next question is who in your building is taking photographs, and that is where wearable cameras come in.

Meta now sells glasses with a camera built in, made with Ray-Ban and Oakley, and other manufacturers sell similar products. They look like ordinary glasses, and they capture photographs and video hands-free, without anybody reaching for a phone.

Most people who wear them do so for entirely innocent reasons, and that is precisely the difficulty. Your staff cannot tell the difference between a colleague who bought fashionable frames and a colleague who is filming them, and neither can your customers or anyone visiting your premises. The uncertainty alone changes how people behave, which is why this belongs in a wellbeing conversation rather than only in a policy review.

It also raises questions that nobody has put to you yet. Can somebody wear these into a meeting, onto a ward, onto a shop floor, or anywhere your staff get changed? What happens to footage of a colleague who never agreed to appear in it? The ICO’s guidance on monitoring workers covers the data protection side, and once again your advisers are the right people to ask about your obligations. What matters here is that somebody who suspects a colleague is recording them behaves differently all day, and will very rarely say so out loud.

Why staff wellbeing data never shows this

That last point runs through everything above. Each of these behaviours is hard to prove, and people do not report what they cannot evidence.

A woman who receives a message at midnight knows the organisation never issued that phone, so the whole thing feels out of scope before she has even decided whether to mention it. A thumbs-down on every message she posts sounds like nothing at all when she says it out loud to a manager, and she knows that before she opens her mouth. Somebody who suspects a colleague filmed her cannot prove it, so raising it feels like making an accusation. And an explicit fake image is close to impossible to report, because there are no comfortable words for describing what happened.

Behind all of those sits one further problem, which is the one that quietly ends reporting inside an organisation altogether. Somebody raises something once, nothing visible follows, and everybody watching learns exactly what that means. After that, your reporting figures look healthy because they are empty, and your staff wellbeing data tells you nothing about what is actually happening.

What HR and staff wellbeing teams can and cannot do

HR teams handle the parts of this they own extremely well. They write the policy, they run the grievance process, they train managers on difficult conversations, they keep records, and they make sure the organisation meets whatever its advisers say it must meet. None of that needs fixing.

The difficulty is that this particular problem asks a different sort of question. An HR manager is not in a position to explain how a group chat behaves on a phone the organisation does not administer, or what an image app can now generate from a work headshot. Nor can they say how quickly something spreads once it enters a private group, or what a pair of camera glasses can and cannot actually do. That is not a gap in anybody’s professional competence. It is simply a different field, and it changes every few months.

What HR can do, and what makes the biggest single difference, is name these behaviours out loud. Staff who have never heard something described tend to assume they imagined it, or that it is too small to mention, or that they will look foolish for raising it. Once somebody stands up and describes the thumbs-down, the midnight message, the fake image and the camera glasses, people recognise their own experience and start bringing it to you.

Two things go alongside that. Tell people plainly that a personal phone still counts as work, because ruling personal devices out of scope does not stop anything reaching your staff, it only guarantees you never hear about it. And tell them what happens next when they do speak, even if the answer is only that their manager will write it down. Most silence comes from not knowing the next step, rather than from any doubt that the behaviour was wrong.

Where training comes in

This is the point at which most employers get stuck. Everything above asks something of a staff wellbeing programme that no policy can supply, because the remaining gap is knowledge rather than process, and it has to come from outside.

Our training programmes fill exactly that gap. A session shows how these apps and platforms really work, and what an image tool can now build from an ordinary work photograph. It covers how unwanted images reach a phone and which settings shut that route down, what camera glasses can and cannot do, and what somebody should actually do in the first hour after something appears. Kids n Clicks also runs continuing support for corporate organisations, because a session delivered in October will not cover whatever arrives in February.

For more on how this behaviour shows up day to day, Kids n Clicks has written about misogyny in the workplace and about cyberflashing, both of which sit alongside everything covered here.

If you lead people, wellbeing or inclusion in your organisation, connect with Parven on LinkedIn or explore our training programmes to book a session for your team.

Parven Kaur is a digital parenting and online safety expert, and the founder of Kids n Clicks. She works with employers on the technology side of online harm and its effect on staff wellbeing. She won the Scottish Cyber Security Award, and she sits on the National Board of Advisors for Barnardo’s Scotland.