Podcast Summary
Podcast Summary for S2E5:My Genetic Counselor Study Notes — Can AI Replace Us?
Yoyo Chu
·
Full episode is available here (in Catonese).
<div id="buzzsprout-player-19298910"></div><script src="https://www.buzzsprout.com/2552094/episodes/19298910-s2e5-ai.js?container_id=buzzsprout-player-19298910&player=small" type="text/javascript" charset="utf-8"></script>
Introduction: The Question That Won’t Go Away
Instead, the episode takes an honest, philosophically grounded look at what the core value of a genetic counselor truly is in the age of AI.
The question “When will AI replace genetic counselors?” has become unavoidable — surfacing at alumni gatherings, academic conferences, coffee chats, and online forums alike.
It isn’t just students or career-changers asking; seasoned practitioners with decades of experience are feeling the anxiety too.
Host Yoyo, speaking from over ten years of clinical practice as a genetic counselor, makes clear from the outset that this episode will not offer comforting reassurance that “AI could never replace us.”
Part One: Automation Was Already at the Door Before AI Went Mainstream
Long before ChatGPT captured the world’s attention, automation tools were quietly reshaping how genetic counselors work.
Platforms such as Invitae’s GIA and Ambry Genetics’ CARE — sophisticated chatbots rather than true AI — were already systematically collecting family histories, conducting preliminary risk assessments, and generating pedigrees before a patient ever sat down with a counselor.
Tasks that once took fifteen minutes per case are now handled by machines.
Rather than viewing this as encroachment, Yoyo frames it as liberation — freeing counselors from high-volume, repetitive administrative work so they can direct their most valuable resource, namely time and focused attention, toward the aspects of care that genuinely require human judgment and presence.
This shift demands an honest reckoning with a deeper question: when machines can handle the fundamentals faster and more accurately than we can, what exactly is the irreplaceable human contribution?
Part Two: “We Understand Ethics” Is No Longer a Moat
For years, a default answer within the genetic counseling profession has been that ethical reasoning and critical thinking are uniquely human capabilities that AI cannot replicate.
Yoyo challenges this assumption directly, arguing that this protective moat is growing shallower by the day.
AI companies have begun hiring philosophers and bioethicists to systematically train models in ethical reasoning.
Complex scenarios familiar to genetic counselors — interpreting variants of uncertain significance (VUS), navigating non-paternity disclosures, weighing the competing interests in pediatric consent — can increasingly be analyzed by well-trained AI with logically rigorous, multi-framework ethical arguments.
“Understanding ethics” remains necessary, but it is no longer sufficient as a professional differentiator. The field must look deeper.
Part Three: Three Distinct Forms of Knowledge — Episteme, Techne, and Phronesis
To map the boundaries of what AI can and cannot do, Yoyo draws on three concepts from Aristotle’s Nicomachean Ethics.
Episteme refers to scientific or factual knowledge — information that can be stated, recorded, transmitted, and retrieved. This is where AI excels: within seconds, a language model can process the entire ClinVar database, draft a structured genetic report, and explain complex hereditary concepts in plain language. In information processing, pattern recognition, and knowledge retrieval, AI has already matched or surpassed human capacity. This, Yoyo says, must be acknowledged honestly.
Techne refers to craft or technical knowledge — the skills acquired through training and practice that govern not just what we know, but how we execute. For genetic counselors, Techne encompasses teachable, learnable, and standardizable competencies: delivering difficult news using the SPIKES protocol, choosing the right analogy to explain penetrance to a patient, or establishing rapport within the first few minutes of a session. AI is closing the gap in this domain rapidly, already capable of simulating standardized counseling dialogue frameworks and generating responses that sound, on the surface, genuinely empathetic.
Phronesis, translated as practical wisdom, is where the deepest distinction lies. Unlike Episteme (knowledge that can be looked up) or Techne (a skill that can be trained), Phronesis is a form of judgment cultivated through lived experience — through repeatedly making real decisions, absorbing their consequences, and reflecting on what happened. It cannot be listed in a textbook or rehearsed in a simulation. It can only be lived into.
To illustrate concretely: Episteme tells you that a pathogenic BRCA1 variant carries a lifetime breast cancer risk of approximately seventy percent. Techne tells you to deliver the information in stages and leave silence after a high-risk result. But only Phronesis allows you to perceive — in a woman whose mother and aunt have both died of breast cancer — that in this particular moment, running through the steps of SPIKES is not what she needs. Twenty years of clinical experience must do the work instead. This is what Yoyo calls “professional intuition.”
Part Four: Professional Intuition — The Signals AI Cannot Hear
Professional intuition is forged through years of real human interaction, and it operates on multiple levels simultaneously.
It is what tells a counselor when to gently press on a question a patient is clearly avoiding, and when to stay silent and let the space do the work.
It is what registers, when a patient says “I’m fine, really” with hollow eyes, that something important remains unspoken.
It also operates at granular, almost invisible levels — for instance, deciding whether to explain a child’s genetic condition to his grandmother using the phrase “an error in the genetic code” or “a small flaw in the blueprint,” and which metaphor will she absorb without quietly turning it into guilt.
This intuition is not simply a communication skill. It is a form of professional sensibility shaped by sustained immersion in particular medical cultures, social contexts, and family dynamics — and because it is rooted in singular human stories and unrepeatable moments, it cannot be extracted from a dataset and learned.
Part Five: Embodied Presence — What a Screen Can Never Provide
When a couple learns that their unborn child has a high probability of a serious genetic condition, every word in the clinical report can be explained by an algorithm — but the weight of the air in the consultation room, the look exchanged between two people, the involuntary trembling of their hands: these are signals that only physical, embodied presence can perceive.
In cross-disciplinary conversations about AI’s limits, two human capacities are cited repeatedly as nearly impossible to automate: haptic and tactile perception, and deep relational communication. In the genetic counseling consultation room, these two capacities are inseparable.
What genetic counselors provide is not only information — it is a form of companionship rooted in shared presence. A screen-based AI can give you answers, but it cannot hand you a tissue when you are crying, hold a silence long enough to make room for shock and grief, or make you feel that someone is facing this with you.
At the most vulnerable and consequential moments in a person’s life, what human beings need is another human being. As Yoyo puts it plainly: “AI has never suffered, so it cannot truly understand suffering. That is why we exist.”
Part Six: A Frank Voice from Reddit
A candid response from a senior genetic counselor on Reddit offered perspective that Yoyo found both wryly humorous and deeply apt:
“If AI becomes powerful enough to take over what genetic counselors do, it has probably already taken over everything else in healthcare. So if we disappear, there won’t be many professionals left sitting at their desks anywhere. Our work is distinct from other medical roles because its core is helping people make important, discretionary decisions. ChatGPT can list all the options — but whether it’s actually good at sitting with a real person and helping them navigate genetic information that is simultaneously abstract, complex, and profoundly personal is still very much an open question. So do the thing you actually want to do. There are good reasons not to go into this field — difficult patients, limited career advancement, the homogeneity of the field, how competitive entry is. But if AI is your only qualm I would not not pursue this for that reason. That’s my two cents.”
Three things resonated with Yoyo. First, the macro perspective: our professional risk has never been isolated — it is part of the broader transformation of all knowledge work.
Second, the honest articulation of what genetic counseling actually is: a profession that exists at the intersection of complex genetic information and the most consequential human decisions.
Third, and most importantly, the closing reminder — that a career should be chosen on the basis of genuine interest and fit, not in reaction to fear of a technology.
Closing: Not to Convince Yourself — But to See Yourself Clearly
Yoyo’s concluding answer to the central question is this: AI will not replace genetic counselors, but it will fundamentally redefine what it means to be an excellent genetic counselor.
The shift required is from “information provider” to “meaningful collaborator” — a navigator uniquely skilled at translating genetic knowledge into personal life wisdom.
The practical implication is clear: let AI handle what it does well — information synthesis, documentation, preliminary case triage — and reinvest the time freed up entirely into what only humans can do well: listening deeply beneath the surface of language, exercising practical wisdom in ethically complex situations, and accompanying a person, or an entire family, through one of the hardest passages of their lives.
In this sense, AI functions as a mirror, forcing each of us to ask: if you strip away the memorized genetics knowledge, the repetitive forms, and the scripted consent explanations — what remains? That remainder is the most irreplaceable asset we carry.
The point is not to reassure ourselves that we are safe from replacement. It is to see, with genuine clarity, where our value actually lives.
For students considering entering the field: this is an era that needs more human depth, not less. For fellow practitioners: embracing this transformation, letting AI serve as a powerful assistant, allows us to concentrate ever more fully on the irreducible human qualities of depth and warmth that no model can manufacture.
Technology will always change. But the human need to be understood, to be accompanied, and to find a way forward through impossible complexity — that will never change.
Next episode: Receiving BRCA positive result: Is this a “bad news”, or a starting point of the new journey.
GENE TALKS Podcast Disclaimer
The content of this podcast is provided for general educational and informational purposes only and does not constitute medical advice, diagnosis, or treatment of any kind.
If you have questions about your personal health, genetic test results, or medical decisions, you should consult a qualified healthcare or genetics professional.
Any cases, experiences, or opinions mentioned in this program reflect the personal views of the host, and do not represent the positions of any medical institutions or professional organisations.
While GENE TALKS and its host endeavor to provide accurate and reliable information, they are not responsible for any consequences arising from the use or interpretation of the content.
Our mission: to help you understand genetics — and yourself — not to replace professional judgment.