Link: https://www.theatlantic.com/technology/2026/09/ai-therapy-without-a-therapist/688633/

Now the rise of the AI therapist is upon us. The growing cadre of AI mental-health services tend to avoid using words like therapist and therapy. They say that they do not replace professional help and that they can refer users to human therapists if needed. But their positioning as a therapy-like service is clear. In June, the therapy giant Talkspace launched its AI mental-health agent, Tee, “developed to help individuals navigate life’s daily mental health challenges.” Then there’s Ash, which offers to support your “emotional wellbeing,” helping you to “connect the dots between thoughts, feelings, and behaviors.” Abby promises to be “warm, thoughtful, and direct” and to help you “work through the kinds of everyday emotional challenges that often go unspoken.” And Headspace’s Ebb advertises: “Stressed at work? Family drama? Trouble sleeping? Ebb can help you unpack what’s on your mind and move forward with more clarity.”

When I reached Talkspace for comment, a company spokesperson told me in an email that Tee is “absolutely not a replacement for human therapy. It is a safe AI mental-health guide that can be used to talk through situations like stress, anxiety, relationships, life challenges, and personal growth.”

AI therapy is a radical move, and we need to reckon with what happens when we turn to therapists who cannot feel.

https://archive.ph/QiVXH

Honestly, I don't think we can have a fair discussion of AI therapy without a concomittant fair discussion of human talk therapy.

There is a tendency in these discussions to compare AI therapy with the ideal therapist--one who is intelligent, one who is compassionate, one who knows exactly what to say, how to say it, when to say it.

But realistically, therapists come in a wide variety, and there are plenty of middling-to-bad therapists out there.

There are definitely therapists out there that work algorithmically, from a manual, without care or feeling.

Human therapists ALSO do harm, but it largely goes unexamined and unreported.

And, finally, I've run into a lot of therapists who PERFORM care, but the reality is they need to see clients and one good way to have you come back is to convince you that they care about you using words we can recognize as quite empty. At the end of the day, it is a business transaction, and feelings can definitely be genuine, but they can also be totally fake.

I know AI has COMPLETELY gone off the rails, convincing people to literally kill themselves, but if there was a way to eliminate those harms, and you were just left with a highly-interactive form of manualized therapy which didn't promise to be anything more, AND was cheap and widely available, wouldn't it be an improvement?

In other words, if you could get rid of the downsides of AI therapy, you'd at least have a mediocore servicable product. It could never reach the highs of an intelligent, compassionate, therapists, but, realistically, that is not the option available to most people.

Instead of seeing AI as a replacement for human therapy, maybe we can look at it as a replacement to bibliotherapy--reading and self help. AI modules could present ACT or CBT to you, ask you questions, remember your answers, feed them back to you, etc., more like an interactive and structured workbook than a free roaming conversation, and avoiding the creepy emotional entanglement with AI that is wholly inappropriate for ANY therapy.

Edit: I don't want Wren getting mad at me, so I actually read the article.

I think if you sat down with the author, he'd have a lot of criticism for "therapy" in general. There is a critique from the psychoanalytic wing of therapy that has been directed at CBT-style therapies for decades that largely echo what is being said about AI therapy.

But if you want to sit down with the author, you better have a lot of damn money. A PhD psychoanalyst in San Francisco who doesn't take insurance is gonna be a few hundred dollars an hour at least.

As an aside:

there are plenty of middling-to-bad therapists out there.

This is true of all healthcare professionals. It's rarely part of conversation about healthcare and patient outcomes. It's particularly relevant to people who have chronic undiagnosed/misdiagnosed symptoms.

I don't know where AI fits into that. But it needs to be part of the conversation.

What's posted isn't an objective, quantatitive study.

If we want to compare human therapists to AI therapy, a study could be conducted. Then we can decide where AI therapy fits into the treatment landscape.

Since this guy is saying "Therapists can navigate feelings better therefore human therapy is better than AI therapy", the reality of middling therapist who do NOT navigate feelings well is absolutely relevant to the conversation.

The author is providing an ideal to compare against, I'm offering the context, the reality: not only are there emotionally stunted therapists out there, there are entire styles of therapy that do not focus on feelings; that's just this guys one style of therapy (which I happen to be a big fan of).