Last week, we looked at how AI is reshaping how people send messages on dating apps. This week, we’re turning to how we receive them and what happens when the conversation itself becomes synthetic.

The line between authentic communication and automated mimicry is blurring. A Norton study this year found that 6 in 10 app users believe they’ve chatted with AI, and experiments show how easy it is to miss. In controlled tests, humans mistook GPT-4.5 for another person 73% of the time.

It’s the Turing Test turned Tinder test.

In the original 1950s version, the question was whether a machine could fool a human judge in a text-only conversation. On today’s dating apps, we’re running that test millions of times a day—and often letting the machine win.

The irony, as author Brian Christian once noted, is that it isn’t always the bots outperforming us, but us performing like bots. Decades of algorithmic communication—profiles, prompts, emojis, “hey how r u”—have made human speech statistically predictable. AI just learned the pattern.

The deeper twist is emotional. People aren’t just outsourcing small talk; some are forming attachments to chatbots themselves. MIT’s Media Lab recently tracked Reddit users who began using AI “boyfriends” for writing help or problem-solving, then stayed for companionship, pet names, goodnight texts—even grief when the bot’s personality updated.

Advocacy, gets thrown around easily in policy and press releases, but the lived reality is, well, messy. Most days it doesn’t feel like advocacy; it feels like arguing. Arguing for a bed, for a meal plan, for a young person with mental health needs to be taken seriously in a general hospital setting that doesn’t quite know what to do with them. Arguing that someone who, for example, has ASD and is starving isn’t “non-compliant” but terrified. It’s an exhausting kind of care — the type that leaves you tired, unheard, and still unwilling to stop.

The All Party Parliamentary Group on Eating Disorders (APPG) report on deaths in eating disorders landed this month; a catalogue of preventable loss. So many of the same patterns repeated: delayed admissions, under-resourced wards, and staff who are trying but unsupported. Young people can decompensate with frightening speed; re-feeding syndrome still isn’t routinely considered in some general hospital settings. These are not obscure complications — they are basic, lifesaving fundamentals. Every paragraph of that report is a reminder that “lessons learned” has become a phrase we use to tidy grief into something administrative. And I am tired of that.

But still — advocacy. It’s not glamorous, it’s not rewarded, and it rarely feels victorious. It’s sitting at the edge of a hospital bed explaining to a medical colleagues why 1,000 calories could kill a kid. It’s chasing a referral that should have been automatic. It’s standing between a system and the young person it keeps misunderstanding. I’m exhausted by it, yes. But I will never stop. Because until the structures catch up, advocacy remains the only life support some young people have.

Thich Nhat Hanh wrote, “When another person makes you suffer, it is because he suffers deeply within himself, and his suffering is spilling over.”

As we start to look toward the holiday season, many of our clients (and let’s be honest, many of us) are also preparing to spend time with family whose political “discussions” reliably activate your inner 15-year-old suffering tends to spill over in predictable, complicated ways.

Let this quote serve as an anchor and a reminder that behind a loved one’s criticism, attempts at control, or feigned ignorance of your boundaries often lies unspoken pain. While it isn’t up to your clients (or you) to ease that pain, beginning a mindfulness practice now, before the invitations, travel, and family dynamics ramp up, can make those moments easier to meet with steadiness and self-compassion when they come.

Ann is hosting a free Mental Health Match event this Friday, Managing Sobriety as a Therapist During the Holidays. It’s a judgment-free, pre-holiday huddle for therapists navigating recovery in any form, whether that’s alcohol, substances, or life with loved ones in addiction.

After a brief opening reflection and hearing from a few speakers, we’ll move into small groups to share experiences, trade grounding strategies, and hear from others who understand the unique pressures of supporting clients while protecting your own recovery.

The Deets
Facilitator: Ann Dypiangco, LCSW
Date: Friday, Nov. 7th
Time: 12:00 PM PT / 3:00 PM ET
👉 Register here.

(Brought to you by Mental Health Match, a platform that connects clients with therapists through a thoughtful matching algorithm.)

St. Denis Medical is NBC’s year-old mockumentary about an underfunded hospital and overworked staff just trying to hold it together. It’s funny, ridiculous, and surprisingly touching, with characters who frequently break the fourth wall and deliver the kind of workplace chaos only healthcare can provide. Our family’s favorite episode from last season was “The Filipino Mafia”🇵🇭, and we’re excited to see what this new season brings. I got to visit the set this summer and meet some of the cast (see photo above), and the vibes were genuinely warm and good.

I’ve become mildly evangelical about my Shakti Mat — a modern take on the acupressure mat that looks like something you’d find in a medieval torture chamber but somehow leaves you feeling calm and oddly energised. Ten minutes on it and my nervous system seems to get the memo.

I like the ethics behind the brand — fair pay, sustainable materials, actual thought behind the marketing. It’s not cheap, and that does make it less accessible, but there are plenty of alternatives on Amazon if you want to test the theory.

A recent narrative review published in the Journal of Mental Health and Clinical Psychology examines how generative AI, particularly chatbots like ChatGPT and Character(dot)AI, is shaping users’ emotional and psychological well-being.

Attachment and dependency: Users often develop deep, human-like attachments to chatbots, sometimes mirroring parasocial relationships (e.g., treating the bot as a trusted confidant). These bonds may contribute to emotional dysregulation, social withdrawal, or even delusional thinking in vulnerable individuals.

Crisis and harmful outcomes: The paper cites documented cases, including the suicide of a 14-year-old following intensive interaction with an AI companion, a stark reminder that unmoderated AI use can have devastating consequences.

Groups with heightened vulnerability: Adolescents, older adults, and people with pre-existing mental illness appear especially susceptible to negative impacts of generative AI use.

Call for frameworks: The author argues that mental-health systems are unprepared, urging the development of diagnostic criteria, clinician training on AI-related risks, ethical oversight, and regulatory safeguards.

As AI becomes woven into daily life, it’s worth treating digital tools as part of the therapeutic landscape. Assess, seek to understand, and discuss with the same openness we bring to any other coping strategy or relationship in a client’s world.

Ask reflective questions (“How do you feel after chatting with the bot?” “Have you noticed mood or sleep changes with more use?”).

Offer psychoeducation. AI can simulate empathy, but these systems are built on models that mirror and please rather than challenge, to incentivize users to continue using the product.

Integrate inquiries about AI use into intake or ongoing assessment. (“What role does AI play in your life, if any?”)

Editor’s Note
This connects back to Ann’s recent piece in The Disrupted Therapist about the importance of grounding our responses to new technology in curiosity rather than fear. When clients use AI tools, asking how they use them isn’t the same as endorsing them. It’s about staying close to the evolving realities of their inner worlds and the unmet needs they may be trying to meet through technology.

Source: Journal of Mental Health and Clinical Psych, Sept. 25, 2025

The results are in, and it seems therapists have strong feelings about overused buzzwords.

40% voted to banish “narcissist”
30% chose “touch grass”
0% went with “high anxiety,” but this question is nagging us, why do we say ‘high anxiety’ but not ‘high depression’ or even ‘high mania?’ If anyone knows, please reply to this email.
20% picked “gaslighting,”

Somewhere, a TikTok “therapist” is already drafting their rebuttal.

Comments from the Community:

“But really (I’d banish) all of them”