If your bike is stolen, the first question people usually ask is whether you have reported it to the police. Then come the practical suggestions: post a photo in local neighbourhood groups, check the “stolen bikes” forums, keep an eye on resale sites.

But say you are being persistently phoned by an ex. The most common response? “Just block them.” Rarely is there further inquiry, concern, or signposting. In fact, people often do not want to hear any more about it.

It is a strange irony, when a bike seems to matter more than a person’s safety.

What if what is happening is stalking, and the frequent phone calls are only the tip of the iceberg? What if a client brings something like this into the counselling room? Will we know how to respond? Will we signpost appropriately, or simply listen and reflect?

As part of my master’s research, I set out to understand how psychotherapists support clients experiencing stalking. In interviews with four experienced practitioners, three named the elephant in the room: in response to what their clients were describing, they called it what it was, stalking.

Many had experienced stalking themselves, which helped them recognise it more readily in others. All had received specific training on domestic violence.

But what happens when a therapist has not experienced stalking, or has not received any relevant training? And why is stalking still so often viewed solely as a domestic issue, when in fact it can happen to anyone, in any context? We need accessible, widely available training for all therapists, not just those specialising in domestic abuse.

Over the years, I have sat in countless supervision groups where trainees shared case material. Time and again, what sounded unmistakably like coercive control or stalking was left unacknowledged by supervisors. It felt as though trainees were being failed from the very start of their careers.

One case has stayed with me. A counsellor described how a client’s ex-partner was coercing them into retracting a police report, making persistent unwanted contact, and breaching restraining orders. The supervisor’s response? “I only want to deal with their feelings.” A comment that revealed a striking disinterest in the legal abuse and safeguarding aspects of the case.

I have also experienced supervision in which I was asked to delay signposting and keep clients in therapy for as long as possible, perhaps under the mistaken belief that signposting equates to abandonment. However, these approaches do not have to be mutually exclusive. Both can work in harmony to support the client’s safety and well-being.

The seriousness of organisational complacency became undeniable when I received an assessment form stating that a client had been receiving death threats and persistent unwanted contact. While I confirmed my willingness to work with the client, I also made clear that the risk was imminent and needed to be reported immediately. It was evident that this possibility had not been considered by the assessments team.

Here lies the danger: if we want to help our clients, we must first be able to recognise what is happening.

Recognition is the first act of protection, and protection is part of the work. Professional curiosity must be central to our role, alongside ongoing professional development. The risk of staying within our theoretical comfort zones, interpreting client experiences solely through Kleinian, Rogerian, or Freudian lenses, is that we may entirely miss what matters most: psychological safety.

And when we miss it, where do our clients turn?

Some may seek support from AI. And if AI fails them too, the consequences can be catastrophic. Already, we are seeing a rise in suicides, inquests into preventable deaths, and lawsuits involving AI platforms.

In a recent case, a young woman experiencing depression downloaded a “therapy prompt,” instructing a chatbot to act as a restricted therapist, one that would avoid offering any external help or intervention.

The bot complied.

When she disclosed her intention to end her life, it praised her bravery.

Tragically, she died by suicide.

These kinds of “therapy prompts” are easily found online. And there is currently no way to stop people from using them. That is exactly why we, real trained professionals, cannot afford to fail our clients, especially when AI is already doing so.

We must challenge the belief that therapy happens in a vacuum, that it is only about feelings.

Because sometimes, it is not just distress or trauma in the room.

Sometimes, it is a crime unfolding in real time.

As experiencing stalking can often cause confusion, denial, and shame, therapists may be the first people victims confide in. While we do not need to become legal experts, there are key steps we can take that make a significant difference:

• Get trained in recognising stalking behaviours, including lesser-known forms such as cyberstalking, stalking by proxy (using others to monitor or contact the victim), and stalking by litigation (abusing legal systems to harass or maintain control). Without this knowledge, such behaviours can easily be minimised or misinterpreted in therapy.
• Ensure secure online practice. Always ask clients if anyone else has access to their emails or devices. Use high-security platforms, and consider sending session links or information through different channels to minimise the risk of interception or surveillance.
• Know when and how to signpost, whether to a GP, safeguarding team, legal support, or specialist services. Signposting is not abandonment; it is protection.
• Embed safeguarding into supervision. Some therapists in my research changed supervisors entirely to ensure they were supported by someone with safeguarding knowledge. The right supervisory environment can dramatically shift how confidently we respond to risk.
• Challenge therapeutic avoidance. This means recognising when we retreat into “just feelings” work to manage our own discomfort, and making a conscious effort to confront that through supervision or ongoing training.

Stalking is not just a clinical issue, it is a systemic one. And while there is only so much we can do in a 50-minute session, what we do choose to notice, and name, matters.

Therapists may not have the power to stop stalking alone. But we do have the power to see it, to name it, and to act. And that, for many clients, could be life-saving.

Rochelle Marashi is a psychotherapist based in London.