TL;DR
A Gottman Institute report argues that AI and video therapy can help people learn and access care, but they do not necessarily capture the full emotional dynamics of a couple in a session. The report emphasizes that a therapist’s observations are prompts for questions, not diagnoses, and says the limits vary by technology and situation.
The Gottman Institute says artificial intelligence and video sessions can support relationship learning and expand access to therapy, but they may not convey the full emotional dynamics a therapist can observe while a couple is together. In a report on technology and couples therapy, a therapist describes how physical cues and the way partners respond to each other can open questions that written accounts or a camera view may miss.
Get gifts for the two of you delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The report’s central distinction is between understanding emotions and experiencing them with another person. AI can explain relationship patterns, suggest communication strategies and offer scripts for difficult conversations, the author writes. But receiving an explanation does not necessarily help partners feel more connected or reveal how each person is experiencing an exchange in real time.
The author recounts working remotely with a client during the COVID pandemic. In video sessions, the client appeared calm while talking about painful experiences. At a later in-person session, the therapist noticed the client vigorously petting the therapist’s dog and tapping her foot. The therapist did not treat those actions as proof of a particular feeling. Instead, the therapist described what was visible and invited the client to discuss what might be happening.
The report says that video therapy can be valuable and that meaningful therapeutic relationships can develop through a screen. However, a camera’s frame may exclude a client’s hands, posture or movements, and may make it harder to see how partners respond to each other. The author also cautions that body cues are not definitive answers: therapists can use them to ask questions, rather than assume they know what someone is feeling.
How In-Person Cues Inform Therapy
For couples, a disagreement often includes more than the words being spoken. A pause, a change in tone, a partner turning away or a visible shift in breathing may shape how each person interprets the other. The report argues that a therapist can observe parts of that exchange and slow it down, helping partners examine both what occurred and what they assumed it meant.
For example, one partner may interpret the other’s silence as indifference. A therapist might notice clenched hands or changed breathing and ask the quiet partner what is happening, then ask the first partner how they understood the silence. Those observations do not establish whether either interpretation is correct; they create an opportunity to ask. That distinction matters because behavior can have different meanings for different people, and a therapist’s role is not simply to label visible distress.
The point is not that digital care is ineffective. The report acknowledges that remote sessions can make treatment accessible to people who might otherwise struggle to receive it. Rather, it identifies a limitation readers should keep in mind when comparing AI advice, video therapy and in-person work: each offers different information, and no single cue or written account necessarily captures a whole relationship.
Popular right now
Samsung 65-Inch Class M70H Series 4K Mini LED Smart TV (2026)
KitchenAid Artisan 5-Quart Tilt-Head Stand Mixer, Almond Cream
Dyson Ball Animal 3 Upright Corded Vacuum
As an affiliate, we earn on qualifying purchases.
From AI Advice to Live Interaction
The report follows an earlier Gottman Institute article titled “When AI Makes Couples More Certain, But Not More Connected.” Its author says that AI may help couples feel they understand why a partner behaves or reacts in a certain way, while that understanding alone may not create greater closeness. The new report extends that question from AI-generated explanations to what therapists observe during sessions.
The author also describes technology as a possible resource for learning and reflection, including a setting where people may examine assumptions. The discussion is not a claim that technology has no place in relationship care. It compares the information a person chooses to provide to a system with the additional visual and interpersonal cues that may be available in a therapy session.
In the report’s account of couples work, a therapist pays attention not only to signs of distress but to how one partner’s emotional experience affects the other and whether the partners recognize and respond to those moments. That focus on the interaction between partners, rather than an isolated description of an argument, is the basis for the author’s argument about what technology may miss.
“There’s a difference between understanding emotions and experiencing them with another person.”
— The report’s author, a therapist writing for the Gottman Institute
What Digital Tools May Miss
The report does not compare outcomes for AI-assisted advice, video therapy and in-person couples therapy, or provide data showing how often video sessions miss relevant cues. Its examples describe one therapist’s experience and support an argument about possible limitations; they do not establish that remote therapy is generally less effective.
The author says systems based on written descriptions rely on information a person chooses to share, which may reflect that person’s perspective. The report adds that tools capable of analyzing audio or video may still struggle to interpret signals in the context of a particular relationship. It does not assess specific products, explain what their systems can detect, or establish how well any one tool performs.
It is also not possible to infer a person’s feelings reliably from one physical cue based on the report. Stillness, tapping or silence can have different explanations. The author’s stated approach is to notice a cue and ask about it; what a behavior means remains for the person to describe and for the therapist and clients to explore.
Questions for Choosing Care
The report does not announce a new study, policy change or technology launch, and it does not set out a timetable for further developments. Its immediate contribution is a practice-based argument about how technology and therapy may differ. Readers considering support can use the distinction it raises to ask what kind of help they want and what the available format allows a therapist to observe.
For people seeking couples therapy, a practical next step is to discuss with a qualified provider whether remote or in-person sessions fit their circumstances and what to expect from either format. The report supports neither a universal preference for in-person care nor a conclusion that AI should replace therapy. It instead suggests that tools offering explanations or conversation prompts may serve a different purpose from a therapist helping partners examine an interaction as it unfolds.
Any broader judgment about the relative effectiveness of these approaches would require evidence beyond the report, including research that compares outcomes and accounts for access, client needs and the type of care provided. Until then, the report’s clearest takeaway is narrower: technology may assist reflection, while a therapist’s observations can prompt questions about a couple’s live interaction, without making those observations definitive.
Key Questions
Does the report say AI cannot help with relationship problems?
No. The author describes technology as a potentially useful resource for learning and reflection. The argument is that information or suggested scripts do not necessarily create connection and may not capture the full context of a couple’s interaction.
Does the report say video therapy is ineffective?
No. It says video therapy can make treatment more accessible and that meaningful therapeutic relationships can develop remotely. The author’s concern is that a camera may limit what a therapist can see, depending on its frame and the interaction.
Can a therapist tell what someone feels by watching body language?
The report cautions against treating physical behavior as a definite answer. A therapist may notice a change, such as tapping or becoming quiet, and ask the person what is happening rather than assume what the cue means.
Does the report provide evidence that in-person therapy works better?
No comparative outcome data are provided. The report offers a therapist’s account and explains possible differences in what can be observed across formats; it does not establish that in-person therapy is generally more effective.
Source: gottman.com
