Josh Briscoe

It's remarkable some people don't believe deception is harmful, even if the deception isn't discovered. Furthermore, deception harms the deceiver by making the kind of person who deceives.

The PPP is grounded on the presumption that a patient with the capacity to make a decision is given information and they yield a response. It treats the patient as a machine, which is why we believe a machine could so easily replace them.2 But that’s not how decision-making works. It’s laden with emotion and values. It’s fraught with negotiation. Most people aren’t sitting at home pondering their...

familymeetingnotes.substack.com/p/a-machine-... Some thoughts on using #AI for surrogate decision-making.

I've never said yes to an interview so quickly as when @ashleybelanger.bsky.social reached out to discuss a topic that – it will surprise no one to learn – I feel pretty strongly about.

Treatment decision making for incapacitated patients: is development and use of a patient preference predictor feasible? - PubMed It has recently been proposed to incorporate the use of a "Patient Preference Predictor" (PPP) into the process of making treatment decisions for incapacitated patients. A PPP would predict which trea...

Bioethicists have been debating the idea for a while, e.g., pubmed.ncbi.nlm.nih.gov/24556152/. Now others are trying to empirically validate the concept. I think it is a case where bioethics for some is really just figuring out how to use technology to support a single bioethical principle.

A Machine Better Than Family Artificial Surrogate Decision-Makers (i.e., the Patient Preference Predictor)

In one sense, AI surrogate decision-makers are the right answer to the wrong question: familymeetingnotes.substack.com/p/a-machine-... #hapc #ai #medsky

Nothing Better than Silicon A Tale of Surrogate Decision-making in Three Parts

A response in narrative: familymeetingnotes.substack.com/p/nothing-be... #hapc #medsky #ai

Should an AI copy of you help decide if you live or die? Doctors share top concerns of AI surrogates aiding life-or-death decisions.

From @arstechnica.com: arstechnica.com/features/202... My answer, as a palliative care physician: never.

this was such an informative read. @emilymoin.com is really good at breaking down the problems here

Looking Again at SSRIs in Adolescent Depression and Anxiety “Society gets the type of adolescent that it expects and deserves.”

Looking Again at SSRIs in Adolescent Depression and Anxiety

“Society gets the type of adolescent that it expects and deserves.”

A response of sorts to @ploederl.bsky.social‘s recent blogpost

www.psychiatrymargins.com/p/looking-ag...

So maybe 15 years from now none of this will matter (though I'm skeptical) -- but in the meantime, I think I stand by what I told Dhruv in that story, if we (meaning educators) don't figure out how to train the next gen in this environment, we're all screwed.

My biggest concern about this technology in the short (and, well, medium)-term is what is does to US, especially the current generation of doctors in training. Even the difference between my third year and first-year residents is pretty stark in their use of AI tools (especially Open Evidence)

What's a Good Death... ...apart from a good life?

...medicine only has a modest influence over whether someone has a good death... familymeetingnotes.substack.com/p/whats-a-go... #medsky #hapc

Page 1 of the editorial "Machine Learning Cannot Replace Surrogate Decision-Makers in Resuscitation Decisions for Incapacitated Patients" Bottom: Read full article at ai.nejm.org.

Editorial by Robert D. Truog, MD, MA, and R. Sean Morrison, MD: Machine Learning Cannot Replace Surrogate Decision-Makers in Resuscitation Decisions for Incapacitated Patients nejm.ai/3Vysymr

#AI #MedSky #MLSky

I don't discount the possibility that AI might be helpful in a number of different ways, but there's very rarely an unmitigated good in health care. There's almost always a trade-off, and we should consider that instead of writing advertisements for journals.

I'm not wholly against using AI in some settings (e.g., I use OpenEvidence occasionally), but we need to count the cost of its use. We can learn from history to discern how AI might impact us socially, existentially, and, yes, philosophically. #medsky #ai

These authors come close to highlighting lessons from our experience with EMRs. It's like their hope in AI overwhelms their realistic appraisal of how the EMR and AI are alike in that they're both forms of technology which shape us. web.cs.ucdavis.edu/~rogaway/cla...

My Machine-Teacher, the Electronic Medical Record Notes from a Family Meeting, Vol. 3, No. 8

If the relationship fundamentally changes, it's because one or both parties in the relationship are dehumanized by the ever-increasing mechanization of the clinical encounter. familymeetingnotes.substack.com/p/my-machine...

4. "Expect the physician-patient relationship to evolve." This prediction is grounded on a wholly technical understanding of that relationship in which the clinician is responsible for providing services to a customer. This relationship is about a person in need seeking help from another human.

Where The Walls Listen Artificial Intelligence and Clinical Scribes

3. The fatalism is obvious. They draw the parallel w/ EMRs but don't draw out the implication: "Just as the electronic health record forced clinicians to spend more time at computers, GenAI will alter how clinicians work." What will #AI force us to do? familymeetingnotes.substack.com/p/where-the-...

Crafting Health Of Workmanship and Medicine

2. Overlook that clinicians must have certain qualities in order to use tools well. It's possible to over-jig health care. familymeetingnotes.substack.com/p/crafting-h...

1. Fit the clinician to the tool (EMR - e.g., make them navigate labyrinthine menus; AI - prompt appropriately).

How Physicians Can Prepare for Generative AI This Viewpoint discusses basic skills for using generative artificial intelligence (AI) and how to take an active role implementing AI into patient care.

It's remarkable to see the same mistakes we've made with the EMR be re-made with #AI, as suggested in this article from @jamainternalmed.com jamanetwork.com/journals/jam... #MedSky

Recovery Through the Storm: A River’s Hard Lesson - Mockingbird One Year After Hurricane Helene

mbird.com/suffering/re... from Aaron McKethan at Mockingbird

Episode #238 ... Frankenstein - Mary Shelley Podcast Episode · Philosophize This! · 10/08/2025 · 30m

I reflect on Frankenstein more than the Kreb cycle. I wonder what that says about #meded. Or me. podcasts.apple.com/us/podcast/p...

I have expertise in both the technical and ethical aspects of clinical AI, so one of my most frequent refrains is that we need to separately address questions of "can we?" and "should we?"

Anyway, absolutely not.

Our greatest concern as clinicians is that the data presented in this article will be interpreted by the media and other readers as evidence that ML can replace surrogates in end-of-life decision-making. “Machine Learning Cannot Replace Surrogate Decision-Makers in Resuscitation Decisions for Incapacitated Patients” by Robert D. Truog, M.D., M.A., and R. Sean Morrison, M.D.

Although ML has been shown to outperform surrogate decision-makers in predicting patient preferences for CPR, the authors of a new editorial argue that ML cannot replace surrogates in making decisions for incapacitated patients. nejm.ai/3Vysymr

#AI #MedSky #MLSky

The sad thing about this is clinicians are so beleaguered by the burdens of the EMR, they'll take any port in a storm without fully appreciating the hidden costs. It looks very official to study things like data privacy but overlook the surveillance risks. #ai #medsky

The evidence I've seen suggests patients do not like their clinicians typing away looking at a computer screen as they attempt to share sensitive details about their health. Presumably ambient AI scribing might fix this, but I have my doubts because clinicians have many reasons not to listen well.

I’m amazed anyone could believe that once an AI scribe saves a clinician time, administrators will leave the reclaimed time untouched. If a clinician could see 14-16 patients in a day without an AI scribe (already a number too high for primary care and most sub-specialties), surely they could see 20-25 patients with an AI scribe. The encounters become shorter because what is expected isn’t human connection but a technical service rendered ever more efficiently by other tools in the AI suite. That reclaimed, unbillable time shouldn’t be given back to those 14-16 patients because presumably those encounters are already optimized. It should be given to 6-9 other patients who can be seen today instead of next week. This isn’t good for the clinicians but the encounter isn’t about them. They are parts in the machine, and may soon be expunged by AI anyway. Ironically, neither is it good for the patient who finds themselves treated as a bureaucratic client, a machine, or an animal, depending on the flavor of dehumanization their condition warrants.