I invite you here to explore six major trends already visible in the psychotherapy landscape. This is not about magical predictions, nor about announcing “here is exactly what will happen”. The idea, rather, is to spot weak signals, movements already under way, and to ask: what does this change for therapists, for patients, for couples, for families, and more broadly for the way we think about mental health? Keeping a finger on the pulse of the present is not clairvoyance. It is preparation, and it is care.

1. The “sex recession”: less frequency, more meaning

For some years now there has been much talk of a sex recession: a decline in the frequency of sexual activity, particularly commented on in Western countries. American figures fuel the debate. According to an analysis based on the General Social Survey, the proportion of American adults aged 18 to 64 reporting sex at least once a week is said to have fallen:

Earlier work had already shown a rise in sexual inactivity among some young adults, notably men aged 18 to 24, between 2000 and 2018.

But this is where caution is needed: a drop in frequency does not mechanically mean a drop in relational quality. Many couples suffer not so much from having “less sex” as from no longer knowing what sexuality means between them.

• For one, making love means “I am still desired”. For the other, “something is still expected of me”.
• For one, its absence means “you don’t love me any more”. For the other, “I am exhausted, I no longer have access to my body”.
• For one, sexuality is a language of attachment; for the other, it has become a place of pressure.

The therapeutic work then consists of reopening the sexual conversation, without shame and without anything normative about it.

In 2026, even therapists who are not sex therapists will probably need to be more at ease with these questions. Because sexuality is not an optional extra of psychic life: it touches the body, the bond, identity, attachment, shame, safety and health.

2. Parental burnout: when the whole family is overloaded

We have been talking about professional burnout for a long time. Parental burnout, for its part, is becoming a central issue, because it simultaneously affects the individual, the couple, the children, the extended family, school, work, the economy, screens and educational norms.

Parents today carry a considerable load of stress: financial instability, the cost of childcare, isolation, worries about their children’s mental health, social polarisation, world events. And this stress does not affect only the parents; it colours the emotional climate in which children grow up — children who notice and absorb what they perceive in the adults.

With possible consequences for the parent’s mental health, for the couple’s relationship and for behaviour towards the child.

This gap is often widened by an exhausting contemporary injunction: one is supposed to be a parent who is available, calm, mindful, informed, kind, stimulating, protective, emotionally regulated, attentive to screens, to school, to food, to the couple, to work — and to oneself. The most solid therapeutic avenue does not come down to parenting strategies applied like a sticking plaster: it begins with the regulation of the parent’s nervous system, the precondition for any possible co-regulation with the child.

3. Artificial intelligence enters the consulting room

A few years ago, we wondered whether patients would consult Google before the session. The question has changed in nature. Today, they are already using AI to put their symptoms into words, interpret their relationships, demand a diagnosis, prepare a break-up, analyse an argument, draft a message to their partner or obtain immediate emotional validation.

And the phenomenon is anything but marginal. A study published in JAMA Pediatrics in 2026, covering American adolescents and young adults, estimates:

For its part, the WHO warned in March 2026 about the growing use of generative AI tools for emotional support, pointing out that many were neither designed nor tested for mental health, and recommending that these uses be treated as a public health issue. In France, the HAS and the CNIL have announced joint work on good practice in the use of AI in care settings — quality, safety, personal data, responsibility.

AI is thus altering the very environment of therapy. Many patients treat it as a kind of co-therapist. Clinicians will need to develop a new competence: not only psychological literacy, but a clinical digital literacy. That means approaching the tool with openness, without shaming the person who looks for answers — that need always comes from somewhere — while setting the record straight when the AI gets things wrong. One can imagine that, in future, we will ask patients to “bring your conversation with the AI to the next session”, just as we ask today for their journal notes.

4. The therapist is no longer the only source of care

For a long time, the therapist could be perceived as the main source of psychological understanding. That is no longer the case. Patients now arrive with podcasts, videos, Instagram accounts, self-help books, newsletters, Discord groups, coaches, paid communities, apps, AIs, influencers — and a whole vocabulary: trauma, narcissism, ADHD, avoidant attachment, hypersensitivity, dissociation, inner child, gaslighting.

This phenomenon has positive effects: it democratises psychological language, helps put words to lived experience, can reduce shame and prompt people to seek help. But it also has problematic effects: clinical terms harden into rigid identities, hypotheses become certainties, nuances fade, and ordinary conflicts end up pathologised. The digital environment is deeply ambivalent. An investigation by the Guardian into the hundred most popular TikTok videos under the hashtag #mentalhealthtips concluded that more than half contained some form of misinformation. Another review, reported by Euronews in 2026, pointed to as much as 56% of often inaccurate or unsubstantiated content in a corpus of 5,000 posts.

Rather than treating these resources as competition, the therapist has much to gain from seeing them for what they are: an ecosystem of actors — the therapist, the coach, the online community, friends, the book club — in which no single player carries the whole weight of change any more. For even the best group cannot accompany someone through a process-oriented intervention, one that follows the nervous system, attends to grounding, to mobilising resources, to pauses, to slowing down, to regulation. That demands a focused individual attention that social networks will not replace: they open a door, but they are no substitute for clinical assessment, context, history, the body, contradictions, slow temporalities.

5. Avoidant attachment: no longer pathologising the partner who withdraws

Attachment concepts have moved out of the specialist literature and massively into everyday language. We now hear: “I’m anxious”, “he’s avoidant”, “she’s disorganised”, “my partner is emotionally unavailable”. That is useful — but these words risk being wielded as weapons.

The more anxious partner is often more in touch with their suffering, and therefore more inclined to read, seek therapy, understand, change. The more avoidant partner asks for less — not because they feel nothing, but because their core strategy consists precisely in not coming into contact with their pain too quickly. From this an explosive asymmetry is born. One ends up saying: “I do the work, I read, I go to therapy, I have understood my wounds. You are avoidant, you make no effort.” Even if part of that is true, the effect produced is the opposite of the one sought: the avoidant partner feels judged, pathologised, attacked — and closes down further.

The decisive therapeutic trend here consists in not confusing accountability with accusation. The first move is to win the avoidant partner’s buy-in by stating clearly that the therapeutic space will not be a place where they are pathologised. Once safety is established, one can lean on the resistance rather than fight it:

We often discover good reasons behind the refusal — an association between therapy and divorce, the fear of being demonised, distrust of a third party. Telling someone they should not see things that way is the surest means of locking them into their position. Conversely, banking on curiosity and the creation of emotional safety — “let’s meet on a human level to talk about our fears and our pain” — is the starting point for repairing just about anything.

6. The return of communal healing

We have long thought of therapy as a very individual adventure: someone suffers, consults, tells their story, understands their patterns, works on themselves. That representation remains useful, but it is incomplete. For a human being never suffers only “inside themselves”. They suffer in bonds, in missing bonds, in groups that excluded them, in families that could not contain them, in institutions that weakened them, in a society where one can be permanently connected and yet profoundly alone.

Hence the return of groups, experiential workshops, therapeutic retreats, talking circles, communities of practice — spaces where people come not only to “understand their problem”, but to have, once again, the experience of a sufficiently safe bond. If our first experience of the group — often the family of origin — was painful, insufficient or traumatic, then part of the repair will probably have to come through another experience of the group. Not only through individual understanding. Not only through interpretation. But through a corrective relational experience. These groups offer a larger container for carrying what is too heavy for one person alone; and something else often happens there — a loosening of the body’s armour, sometimes laughter and play, the rediscovered experience of the pleasure of being together.

The need is tangible.

Therapy is going to have to relearn how to work with belonging.

Conclusion: therapy is becoming more systemic, even when it does not say so

Taken together, these six trends trace a clear movement: the therapy of tomorrow will be increasingly compelled to think in systems — even when it works with a single person.

Because a person is a body, a history, a family, bonds, missing bonds, institutions, belongings, loyalties, algorithms, norms, material constraints. A person is a being of stories, moving through a world saturated with dominant, normative and unsettling narratives.