Welcome back to our conversation with Elizabeth Cotton, author of UberTherapy. If you have not listened to Part 1, I recommend starting there.

We begin with Cotton’s own journey through psychoanalysis and labor organizing and her experiences engaging directly with digital mental health platforms. We also talk about what it can feel like to try to reach a real person through the architecture of app-based therapy, where its consumer logic can make connections harder to find and trust.

In Part 2, we zoom out to the systems that give us these experiences. Cotton traces how public austerity and platform capitalism have combined to turn mental health care into a set of digital products, governed by algorithms, data extraction, and dynamic pricing. In this world, qualified human therapists are slowly displaced by AI-driven “solutions,” while those who remain are pushed into precarious, low-paid platform work.

But the conversation also turns to what we avoid and miss out on when care is built to be fast, standardized, and scalable: our tolerance for dependence, our ability to sit with loss and limitation, and our willingness to stay with another person when the work gets messy. We close by asking what “real therapy” might require, and what it could look like to protect the conditions that make it possible.

The transcript below has been edited for length and clarity. Listen to the audio of the interview here.

Justin Karter: You talk about how the marketing around these forms of therapy is that they’re highly personalized, right? Part of how it gets set up is that, by using your data, they can market to you exactly what they think you might need, or something like that. But it’s actually a really depersonalized form of therapy that’s on offer. I love the way you say this in the book. The quote is that “you can have everything you want as long as it’s iCBT light,” which you call “a cheap and politically convenient model of therapy that defines the problem at the level of the individual and their cognitions and behaviors.”

What is it about this particular form of therapy that’s becoming ubiquitous? Why is it the only form of therapy being offered on these platforms? How does this form of therapy actually prevent the sort of space that you’re talking about—a space to speak your mind—from occurring?

Elizabeth Cotton: I’m going to talk about the United Kingdom to answer this question. I think it’s still relevant to a United States audience, because in a way, all roads lead to Rome.

In the UK, what we’ve seen in the NHS is a profound standardization, manualization, and industrialization of therapy since 2008, when NHS Talking Therapies, it used to be called IAPT (Improving Access to Psychological Therapies), Increased Access to Psychological Therapies, was introduced. We all got excited about it. More people are having therapy for free in the UK, a beautiful thing, loved it. The next generation of therapists leapt into it. Lots of training was offered.

What emerged over time, with lots of money, was a short-term, standardized, solution-focused model of therapy. CBT, increasingly iCBT, but very linked to welfare reform and austerity. It started in 2008, when everything began to go really wrong for this generation of users and therapists.

What we saw was a real co-option of mental health services into welfare reform. And the fundamental shift, which we saw right from the beginning but just didn’t want to recognize, was the use of psychotherapy to solve political and financial problems.

Demand goes up, supply doesn’t, not enough, so we’ve got to make it cheaper. It replaced a clinical logic with a financial logic. That sounds simplistic. It’s just a fact. That’s what was driving it. It was the argument that if 50% of people recover after six sessions of CBT light, then 50% of the benefits bill, the welfare benefits bill in this country, will be cut, because you’ll have a huge percentage of people who are off sick for mental health problems going back into work.

Under this logic, work became a health outcome. The outcome, the measurement of whether we’re providing a good service, was how many people can come off benefits and welfare and into employment, regardless of what that employment was. We had decades in this country of the “work is good for you” mantra, which, if you’re working as an Uber therapist, you’ll know is absolutely untrue. It’s disastrous.

Every survey I’ve ever done of therapists in the UK 70% are depressed and burnt out. It goes even higher when you enter these highly industrialized systems. And in a way, it doesn’t matter.

That’s another argument in UberTherapy. This isn’t about therapy. This is about digital health companies, and not even digital health companies, just digital companies. This is about making money. This is about cutting state expenditure and increasing economic activity and making money, partly at certain points through datafication, through the amount of data you could capture from the NHS.

I think, how did this happen? It happened by stealth. We didn’t know this was the endgame. We didn’t understand that the shift in logic would have such catastrophic effects.

In the book, I talk about one company, because I go to all these meetings about psychotherapy, health, and policy. I remember this company in the early days, 15 years ago, becoming one of the first and largest digital providers in the NHS and IAPT. They started out quite well, in that they were basically offering the same therapy but online. But over time, they began developing manualized and digitized services that now come in.

Most importantly, until five years ago, this company would literally put on its website, and go to conferences, and say: “We have captured data from a quarter of a million NHS therapy sessions in order to develop AI tools.” They literally had it on their website.

We were incredibly naive, as it was considered a good thing. No one asked: hang on, did anybody give consent? How did that happen? What’s the end game of this? What use is that to the NHS, to have you data mining all of those sessions?

Now, in the UK, they are the biggest provider of AI diagnostic tools to both the public and the private sector, based on a decade’s worth of NHS data that no one knew they were capturing. I don’t name the company because I don’t really want to get sued by them. But this isn’t that sophisticated, how this happened. This happened because there was a strategic business intention to do that.

John Pring is a journalist who runs a disability news service and wrote a really important book called The Department, which came out last year. I was so grateful to him because he’s a brave person, because he, like me, went to all of the meetings. But he names names.

And what he shows is, again, it’s not that sophisticated. In the early years of austerity in the UK government, who turned up at those meetings? American private medical insurance companies. They were present in those policy decision-making rooms, and the model conveniently fits their expansion, which is what’s happening here in the UK.

We’re seeing a rapid emergence of private medical insurance in a way that we’ve just never had here before. Also, the emergence of large-scale platforms providing employee assistance programs, which was always a minority player in this country, but is growing and growing, particularly in the public sector, also in universities and schools, so student assistance programs as well.

We’re seeing it come to life now, the intentional and huge expansion of private therapy platforms in this country. But that was determined 15 to 20 years ago. It’s just that we’re seeing it come into action now.

 

Karter: It’s striking to me that if you were trying to create an AI therapist trained on “good therapy data,” one of the worst things you could do is train it on data that’s ranked by very questionable recovery endpoints. Like you said, endpoints that assess whether someone can internalize a CBT framework and use it to return to work. That’s based on data from a form of limited therapy over six weeks, with therapists operating in ways that minimize the therapeutic relationship, to minimize dependence, to make it easier to terminate after a short period of time. Then you give all of that data to an algorithm to figure out what therapy is, and you let it operate on its own with patients.

It seems like the priors built into that system are actually priming the algorithm to do something anti-therapeutic, even if it’s efficient for the sake of returning people to work, for welfare reform, or something.

Cotton: I think on every level, how could this work? Is this a big conspiracy theory? No, I was at a lot of those meetings. I can tell you now that it really did happen. But in a way, it doesn’t matter because it has happened, and we can’t undo that.

On a very personal level, we have people going into a therapeutic system that thinks it’s okay to use standardized questionnaires whose value is hotly contested. At the end of every single session, you’ve got your lovely little NHS therapist going, “Is everything okay? Are we feeling better?” It’s so manipulative. Then, apparently, 50% of people say they’re feeling better.

And then if you’re not, what are you expected to feel? I mean, it’s total manipulation. It’s not a basis for being able to say what you think. For someone like me, it’s a non-starter.

That’s where the damage is. It’s not just on the macro level. It’s on the micro level, too, of what it means for someone to go into that system, being told that this is therapy, and then feeling that they’re being completely manipulated. And either, at best, you just don’t go back, which happens a lot in the NHS, or you feel that you’ve got to spend your time cooperating with somebody who’s fundamentally not able to reach you at a level you know is genuine and authentic.

 

Karter: Let’s talk a little bit about the conditions of therapists on these platforms and in these systems. You highlight the precarity of these Uber therapists, and you write that “they’re the growing majority of mental health practitioners now, working in non-clinical and generic roles that are often unwaged or on low pay.”

While there are, of course, bad therapists, you highlight how it’s really the working conditions and the systems that these therapy workers are caught up in that produce a diminished form of therapy. And in some ways, because the therapy is so diminished, it can create conditions under which AI therapy becomes preferable for some consumers.

Can you lay out the working conditions? What do they look like? What’s it like to live as a mental health worker who’s employed by, or within, one of these platforms?

 Cotton: I’m going to answer in more generalized terms, because at the moment in the UK we don’t have a majority of what I would call Uber therapists, in the sense of working for one platform company. In the UK, it’s more of a portfolio. People will work for a platform for a bit, some will work for EAPs, a bit of part-time work, and a bit of private practice. But I call it the therapy factory because we’ve entered an industrial model of precarity.

When I talk about Uber therapists, I’m talking about entering this grey world of insecure, intense work, because the gig economy intensifies it. You’ve got to take every gig. You’re never sure who’s going to come and who’s going to go. There’s no long-term perspective in this model.

I conducted a survey in 2019 looking at NHS workers across different sectors, but the striking finding was this very intense iCBT light, CBT light model in the NHS. It’s been quite difficult to get it published, but we’re nearly there now because it’s critical. One stark statistic was that 44% of people working within the NHS, across different contractual arrangements, part-time, self-employed, all sorts, said they’d been asked to manipulate performance data. There was a loss of integrity from working within this gaming data system.

Also, 70% said they were burnt out as a result of their work. You’ve got 70% burnt out, 44% who’d been directly asked to manipulate performance data, and yet 62% said they felt they were providing good care. That figure blew my mind. It just doesn’t add up for me.

I thought about it long and hard, and part of me, the “bad therapist” part of me, was like: you’re delusional. How can you say you’re burnt out, you’re gaming your performance data, and you’re providing good care? But of course, therapists are therapists. They work within these systems to the best of their abilities until they can’t anymore. Then they have to go and find other things. Lots of people choose to work part-time as a way of managing the stress, and they lose money as a result.

It made me think about the complexity for therapists, as opposed to Uber drivers. You’re faced with a constant demand to self-harm, to compensate for a broken system through your own energy, diligence, and professionalism. I don’t want to add to the denigration of Uber therapists. I think it’s a bad situation that we have to understand.

And therapists will be asked to be discerning about how and where they work: the cost, whether it’s worth it, the ethical compromise, the professional compromise, and the compromise to their mental health. I don’t advise people to walk out of work. If they haven’t got an alternative, I don’t advise anything that would make their situation more insecure. But this book is an invitation to think about the long-term consequences of this.

The fact is, this is already an unsustainable professional model for most therapists. People duck and dive, doing what they can. But over the long term, this is becoming the dominant model in the profession. It’s important to recognize it for what it is and make strategic choices. And, of course, many people are moving out of therapy. In this country, a lot of people don’t earn their primary income from therapeutic work. They’ll have other jobs, in higher education, teaching, whatever.

I think there’s a real problem in the therapy profession: the huge difficulty talking about class, privilege, and money. In the book, I use the psychoanalyst Sally Weintrobe’s term for climate change: Noah’s Arkism. It’s the idea that enough people are privileged enough to think they’ve got a place on the boat, so they don’t have to worry about the animals that don’t.

In the chapter titled “Do You Have to Marry a Rich Man to Be a Therapist?” I talk about the professional architecture that’s failing to address this. I call it a 17th-century French court, where there’s a refusal to confront class and privilege in the profession, right throughout the food chain: training, accreditation, all of that machinery. It forms what I call a culture of professional cannibalism. We eat each other.

In the UK, we’re going through a process of regulation. We’re so far behind the US, but we’re doing a really bad job of it because we’re establishing a distinct, inefficient, and counterproductive class system within the profession.

Maria Albertson, who ran Counsellors Together UK, was like a Linda Michaels counterpart in the UK. She was part of the Digital Therapy Project. She ran the largest working-class network, Counsellors Together UK. Sorry, my voice is emotional, because she was a good friend of mine. She died a year ago, too young, and our professional culture is at a huge loss because CTUK went with her. The people didn’t, and the ambition didn’t, but she was a huge driving force in raising questions about how the profession would defend itself.

She and I did lots of research projects around the financial landscape. It was really difficult to get people to talk about money, so we used this provocative statistic: 4% of therapists in the UK use food banks because they just can’t earn enough to eat. And whether it should matter to you as a consumer of therapy whether your therapist uses a food bank.

Without political engagement from the key actors, this isn’t going to get any traction. And I’m aware of how weak we are, again, with political leadership on this issue within the profession. It may be that the political leadership will come outside the profession. Maybe we’ll join the Uber drivers’ union. Maybe we’ll have to find other political spaces to collectivize.

What Uber Therapy does is challenge the profession’s defenses around class, privilege, and money. And in Uber Therapy, it’s not just that I don’t think we have a place on the boat. I think what Uber Therapy says is: there is no boat.

 

Karter: There might already be this sort of class system, ICBT-lite digital therapy for the masses, and maybe depth psychotherapy for people who can pay privately, and can pay at such a rate that those therapists can afford to stay outside of these insurance and digital platform systems. But that would be a small boat in terms of who can access that and who can provide it.

I want to move to something else. You hit this existential register. You’ve talked about it a little bit already in UberTherapy, but I found it so compelling.

Cotton: I don’t know why. I haven’t worked out a way of not apologizing for that. I’ve been apologizing for it my whole life. But we are therapists, so we can afford to stare into the abyss a little bit.

Karter: You do this larger existential, cultural critique. You’ve mentioned the defense against thinking. This is part of, actually, let me read from your book, because I think it really hits you in the gut.

The emphasis is that this larger apparatus of Uber Therapy is one of the ways that our cultural moment, the way our society is structured right now, prevents us from thinking about each other and the messy parts of being human.

You say: “Therapists will become just the people behind the platforms, the click workers surviving on predatory prices and wage theft. You will pay a dynamic price, unsustainably low, but seduced by the offer of avoiding a day spent haggling with your insurance company. If you can live with the guilty secret that your affordable therapy is likely to pay your therapist below minimum wage. The demand for continuous growth in the platform economy will, in turn, influence the nature of the therapeutic relationship, where patients become ‘homoconsumens,” quoting Fromm, “demanding a restless and insecure and ultimately indifferent on-demand consumption of therapy that offers a flight from the facts of life.”

What sort of existential avoidance do these platforms offer? What flight from the facts of life? And what does this continual, and perhaps even exponential, avoidance mean for our future of collective organizing, of political resistance?

Cotton: Should we step back from staring into the abyss?

Gosh, you see, this is where psychoanalysis, for me, and the great psychoanalytic writers, really are a balm to my soul. Because once you’re at this point, staring into the abyss, you can’t just cheer up. You really have to find a deep solution to it.

Karter: I laughed when I read you call the NHS version of therapy the “cheer up, love” treatment.

Cotton: Oh my God. I live in a place where people still say that to each other. I’m a menopausal woman. I’m just like, God, if you think I’m not angry now. It’s very triggering.

Karter: Just say “cheer up, love” to me.

Cotton: Yeah. Just cheer up.

I use, in the book, Roger Money-Kyrle’s The Facts of Life. It’s a book I read while studying at the Tavistock and Portman NHS Foundation Trust, when I was desperately trying to find some comfort in being myself and feeling really awkward about seeing everything in such political and existential terms. Because therapy can be a very normalizing profession, and our cultures can be terribly normalizing.

I’m that person, I’m hypervigilant, I’m PTSD. I’m like, yeah, but there’s something happening over there. Can we just, can you get more anxious, please?

Money-Kyrle, I just love the confidence and the broad brushstrokes. He can talk about the whole world, and it makes sense. He can talk about the facts of life.

I reframe them a bit. They’re like my touchstones, my psychoanalytic touchstones throughout this book, of me getting really anxious and upset and staring into the abyss, and then pulling it back to: what is it that we are all dealing with and working with as human beings, the human condition?

We’re not the centre of the universe. We’re not the key player. We came from parents who didn’t even know who we were. We were created by other beings. We’re completely dependent on other people. This is a fact of life. And it is horrific and awful. Most of us would like to avoid that at certain points. And then you die.

In the book, I’m quite playful about the idea of doing everything for as long as you want, in whatever way you want. But actually, there are limits. And there are huge losses that are never addressed through a consumption model.

It really challenges the idea of the consumer king: you can get anything you want, and you can have it right now. If you don’t like it, you can get some other stuff, which is capitalism for you, isn’t it?

The facts of life, the things we always have to return to and wrestle with, are fundamentally not going to be satisfied by this model. I call it psychic pilates in the book.

I feel there are seductions and attractions to having an AI therapist. In the book, I give a brief scenario of how comforting it could be to be in a room with your AI analyst, who shows you what you need to see, without being jarring or challenging, and who doesn’t fall asleep in your sessions unless you want them to. Then you can go on a complete self-righteous rage with them. It helps you feel like you are in control of something that you are not actually.

I suppose that’s the play we have to get comfortable talking about, in addition to the idea that free association requires freedom of association. You should care if your therapist is using a food bank. I mean, if you have an AI analyst, you don’t need to worry about that. You don’t need to worry about whether they’re okay. That might be quite nice, actually. That might be quite a playful tool you could use within the context of therapy.

But actually, as human beings, if we want to touch that existential problem, we have to get involved in dependency. We have to get involved in loss. I would rather navigate that with an empathic therapist, a person I have grown to trust and rely on, than try to find a quick fix for that.

 

Karter:  This feels like a good transition to the “real therapy” project at the end of the book, because you do offer a roadmap. Thankfully, there’s some hope at the end of this, right, after this digital therapy hellscape.

Can you walk us through what you mean by “real therapy” in the book? Where do you see promising examples of resistance and alternative models right now, whether that’s in unions, community projects, or small pockets of practice? Where are people pushing back against Uber Therapy in their own workplaces? Where might we contribute? Where might we start, if we want to protect this idea of real therapy?

Cotton: One of the characteristics of political books, certainly industrial political books, is that you’ve got Chapter Six, which is two pages of solidarity. Let’s just have solidarity. I really didn’t want to do that for this book. I’ve written those books as well. I’ve written a lot about solidarity. In the abstract, it sounds great.

But the reality is that most, even the best activists, are burnt out right now. And we’re resistant to being asked to do yet another thing, comrade. I spent a lot of time on Chapter Six. It’s the biggest chapter, and it’s the one I think is the least formed, in the sense that it’s my work for the future. I’ll tell you as far as I’ve got, and then let’s see if we can join each other.

I wrote it as a series of intentions because things are changing so rapidly, both politically and institutionally. We’re in an age of institutional collapse. Things go, people go, really fast. It’s hard to plan this.

I do ultimately think it’s about relating. I have two personas I’ve used in my work around surviving work. I’m by nature a survivalist, which I would say is anti-psychoanalytic, and I sometimes call myself Sarah Connor, John Connor’s mom in the Terminator movies. I’ve got everything packed. I’ve got a jeep and a hideout. I’m a single parent as well. My mantra is often just: don’t have a stroke today. I’m at capacity.

And I recognize that, in our profession, many of us are middle-aged women. A lot of us are awesome, holding the universe in our bare hands. And we’re being denigrated and demeaned. We’re considered invisible and useless in this digital world, which I obviously contest. So I don’t have a “come on, kids, let’s just do it” response to this. I have a very tired, ancient, and feminist response to this question.

But I also have a grown-up part of myself. I’m in awe of and admire my profession and my colleagues in the therapy world. Of course, there are bad therapists. But I don’t even know why I said that, actually, because it’s not six of one and half a dozen of another. The vast majority of therapists I know are older, experienced, empathic, overqualified, remarkable people. It’s remarkable what we do, and the ethical space we hold.

I start from that, and I’m always surprised by how many of us feel like we’re failing, because we’re set up for failure in these systems.

I would be a really bad Uber therapist. I just couldn’t hack it. I couldn’t be nice in that intense way. So my grown-up answer is: hold the space of self-worth, and worth for our profession. Hold the authority of our experience. Put that before us and prioritize it.

At the very least, it means: if you feel you’re being denigrated in a professional or work environment, step back, walk away, and don’t get busy trying to fix everyone. I know when I go into hyperactivity, I’m compensating for something I’m actually really uncomfortable with.

In Chapter Six, I lay out a list of intentions about our responsibility to hold spaces, to promote thinking on difficult issues in the way we know how, and to be discerning about who we do that with.

Some of the intentions I address are around the political fault lines, because this isn’t just about therapy. This is about politics, and it’s about our culture.

One is defending free speech. I use the example of academia, but particularly in therapeutic research, we’ve got a problem: critical stuff is hard to publish, so those debates don’t happen. In the digital therapy sphere, I’m not saying all the research is bad, and I’m not criticizing individuals. I’m saying financial interests show up in what gets published. Let’s take that on the chin. And I think it’s going to get worse in a world of AI publishing. The truth is often very hard to find.

That means we have to prioritize lived experience. We have to prioritize service user experience, public interventions, and public engagement. We know the right voices and experiences are not being heard in this debate. We have to keep going with this.

Another issue is work-related suicides, which in the UK is not recognized as an actual thing. That means it’s not monitored. It’s not regulated the way it should be. There’s a big fight ahead to understand the harms of work, the harms of digital work, and algorithmic harms. For example, Uber drivers, there are huge legal cases around material and non-material harms. Therapists will have to engage with that. That will be a big part of my work over the next few years: how do we quantify the depth of algorithmic and AI harms for Uber therapists in the future?

We also have to challenge the technology, particularly around surveillance in psychiatric settings in the UK. There are a number of cases that are being poorly handled by the profession. We need to address the harms to service users caused by the technology.

But the underlying ethos is: free association requires freedom of association. Our conversations can’t just be about technique and different forms of psychotherapy. We have to go for the jugular around money and working conditions, and make that an accepted part of professional debate.

In my formulation of “much better help,” I talk about a digital therapy Kitemark as a long-term objective. I’m an associate professor of responsible business. I worked in a global trade union in the mining sector. I’m not a great believer in corporate social responsibility. I’m made of harder stuff.

But I do think planting a flag around a digital therapy Kitemark could be a good way to organize consumers to collectively set the parameters: what we find acceptable, what we don’t, where the harms are, where the risks are, and what the costs are.

Because what we will see is more consumer pushback in the courts. There will be more cases of harm from digital therapy played out legally. Running alongside that should be an invitation to a collective, public debate about what would make a good digital therapy intervention. That has to be held at a cultural level, because we can’t keep throwing bricks at each other and blaming individuals.

We have to set the parameters of what’s worth it. What costs are worth it for this convenience? If they’re not, what would a good model of digital therapy look like?

 

Karter: This feels like a kind of time capsule at the end, because in the book you say a number of times that we’re at risk, because Uber therapy is becoming so pervasive, whether you engage with it directly or just through its influence on the discourse, that we could lose sight of the accumulated wisdom of what therapy is, what therapy could be.

You capture it briefly and beautifully in several places in the book. What is that? What is good therapy? What is much better help? You mentioned before the sort of therapy that helps us feel like other people are worth it. As we draw to an end here, could you remind us what that looks like? What is therapy beyond Uber Therapy?

Cotton: Oh, we’ve been talking for ages, so I might have a little cry. Despite being Sarah Connor, I’m also a complete hippie.

The thing that my therapy, my interactions with therapy, have done and continue to do is the relief of not having to pretend. Not starting from the premise that I don’t fit, or that I have to fit someone else or something else in my environment. I’ve been very slow in my life to hear the invitation to be myself, or to think that that’s a good idea. It’s one of the reasons I write, why I’m funny, and why I make such an effort in my relationships.

There’s a part of me that thinks there’s something really not likable about me. And it wasn’t until I was, genuinely, and often not verbally, in the presence of someone who didn’t ask me to make them happy, or to please them, or not be challenging, or not be rude, and not be Hair-n-Teeth, not show them that part of myself, that over time it became possible for me to sit with those parts of myself that I couldn’t accept.

It’s not news, but self-acceptance and self-love is the hardest journey. And to sit with myself as I really am has taken me a whole lifetime, and I couldn’t have done that on my own. I could never have experienced it.

I’ve had various analysts and therapists who’ve done it in very different ways, and have represented the problem to me in lots of different ways over my lifetime. But what they all have in common is their genuine acceptance of me as I am, and a genuine invitation for me to show myself as I am.

I didn’t necessarily know it at the time, but that’s what they taught me: how to do that. And not to worry about recovery or finding a solution to a particular problem. I was never able to articulate what my problem was. I was never that zipped up. I was never that kind of  “I just want this, and I want to do it in six sessions.” I don’t recognize that in myself.

What I was given was a space of non-judgment. And that’s the only way I was able to say what I was thinking, and what I felt, and who I actually am.

 

*

You can read Part 1 of our interview here: https://www.madinamerica.com/2026/02/ubertherapy-and-…eth-cotton_part1/

 

**

MIA Reports are supported by a grant from Open Excellence and by donations from MIA readers. To donate, visit: https://www.madinamerica.com/donate/

 

The post UberTherapy and the Enshittification of our Relational Lives: Part 2 of our Interview with Elizabeth Cotton appeared first on Mad In America.

 

IPAK-EDU is grateful to Mad In America as this piece was originally published there and is included in this news feed with mutual agreement. Read More