min read ​Read online​

#31 | How To Open The AI Can Of Worms With Clients – Part 2 of 2

Hi there, this is Yash again!

New name in this newsletter (I know), so here’s a refresher: I’m the content and outreach guy at TinT, while also a trainee therapist at TISS, Mumbai.

It’s my last few weeks as a trainee, and I’m cherishing my days here, as I sit down to write this week’s edition.

*Last time, we talked about the first side of this conversation: how to explore your clients’ use of AI without minimising or skipping over it. If you missed Part 1, here’s the link. I’d recommend going back before reading this one. It lays the ground.

This week, we turn to a side that tends to make therapists a little more uncomfortable. Not because it’s complicated, technically. But because it’s personal.

We’re talking about your use of AI. And what does that mean for the people sitting across from you?*

Before You Say Anything To Clients, Have You Sat With This Yourself?

This is where I’d ask you to pause before reading further.

Because how you talk to clients about your own AI use will depend entirely on how clearly you’ve thought it through for yourself. And many of us haven’t. Not really.

A few questions worth sitting with:

• Are you comfortable with clients using AI tools alongside therapy? Does your answer change depending on the tool, the client, or the presenting concern?
• What changes in your practice when you use AI? In your workflow, yes, but also in your sense of the work itself?
• What might change in your client relationships if they knew?

Your stance will shape the conversation. Clarity in yourself will shape how you communicate it. This isn’t about having perfect answers. It’s about not walking in underprepared.

What Are Clients Actually Wondering About?

Clients may not always ask directly. But they might be thinking:

Does this tool have access to my personal information? Where is my data going? How does any of this help in therapy, exactly?

These are reasonable questions. And they deserve answers that are simple and relational, not technical.

Not: “The platform uses AES-256 encryption and does not retain session data beyond 30 days.”

But: “I use a tool to help me structure my notes, and your personal details aren’t shared with it. I’m happy to walk you through what that looks like if it’s useful.”

The difference matters. One is a data policy. The other is a conversation.

Did You Ask If It’s Okay?

AI use in therapy is not just a workflow decision. It can be a relational one, especially for clients who care about any AI use and privacy.

And like most relational decisions in the therapy room, consent matters. You might say something like:

“I sometimes use tools to support documentation or session structuring. Would you be comfortable with that, or would you prefer I don’t use them for your sessions?”

Be prepared for either response.

If a client is uncomfortable, that’s not a problem to solve. It’s information. Consider offering alternatives, limiting AI to non-identifiable data, or simply not using it for that particular client. The therapeutic relationship takes precedence over workflow convenience. Always.

What’s The Ethical Layer Here?

Clients can reasonably ask: Is my data being used? Is anything being shared externally? Is this shaping how you understand me?

Even when the answer to all of these is “no,” opacity erodes trust. And if the answer to any of them is “yes,” transparency becomes even more important. Not as a legal obligation (though that too), but as a relational one.

You don’t need to over-explain. But you shouldn’t obscure either. Something like:

“I sometimes use tools to help with structuring notes or materials, but your personal information isn’t shared. I’m happy to answer any questions about this.”

That’s it. That’s often enough. Clarity builds trust far more reliably than comprehensiveness. This clarity will develop when you know how AI handles client data and privacy (which we have explained in a previous newsletter edition, too).

The Third Side: What About the Therapeutic Relationship?

Here’s the part that gets talked about least.

AI isn’t just something clients use on their own, or something therapists use in the background. Increasingly, it becomes part of a shared ecosystem. And that ecosystem needs conscious navigation.

Some questions worth exploring with your clients, when the moment is right:

• When is AI helpful for you, and when does it feel like it gets in the way?
• Do you want to bring what you explore with AI into our sessions together?
• Are there things you’d prefer to process only here, without AI in the loop?

These aren’t rules. They’re co-created boundaries. And co-created boundaries tend to hold better than imposed ones.

Has The Therapeutic Frame Changed?

AI may be sitting at the base of therapeutic frame today. How much we tend to notice is the question.

AI may be sitting at the base of therapeutic frame today. How much we tend to notice is the question.

For a long time, therapy was a dyad. Client and therapist.

It is increasingly becoming something closer to a triangle: Client 🤝 AI 🤝 Therapist. And that triangle needs to be navigated consciously, because ignoring it doesn’t make it go away. It just means it goes unaddressed.

AI is changing how clients reflect on their own experience. It’s changing how therapists work. It’s changing how meaning gets constructed between sessions. None of that is neutral, and none of it disappears if we don’t talk about it.

So What Is This Conversation Really About?

More than technology, this is a conversation about process. About where and how your clients are making sense of their lives, and what that means for the work you do together.

You don’t need to become an AI expert. You don’t need to have opinions about every new model or product. But you do need to be able to talk about this without discomfort. Because the next version of therapeutic competence may not just be understanding your client. It may also involve understanding the systems that shape how your client understands themselves.

That’s a larger ask than it sounds. But it’s also a familiar one. Therapists have always had to account for the world the client comes from, not just the self the client presents.

AI is, from now on, part of that world.

---

We’ve put together a resource on AI boundaries for clinical practice, if you’d like something concrete to take away from this two-part series.

👉 Explore the resource here​

---

If this piece made you think of a colleague who might find it useful, please do pass it along. These conversations go further when more clinicians are having them.

Wishing you a full Calendly this week,​
Yash
Content & Outreach Lead, TinT
​

Follow along on @be_tint​
For more resources view the website​
Connect with me, Harshali on LinkedIn​