91% of the top medical journals with an AI policy forbid reviewers from uploading your manuscript to a chatbot. That rule protects your confidentiality. It says nothing about you running your own draft through AI before you ever hit submit — which is exactly what you should do.

A reviewer opens your manuscript at eleven at night.

She is an expert.

She is…

Read more

About one in three American adults now ask an AI chatbot a health question in a given year.

That is roughly the same share that turns to social media for health information. So a quiet research question has become an urgent one: when should a medical AI keep its mouth shut?

A new review in npj Digital Medicine pulls together the young science of “abstention” — a technical word for a simple idea.

Subscribe now

It means knowing when to stop, ask for more information, or just say “I don’t know.” The finding is blunt. Today’s models almost never do it. In one study, a leading model held back on only 1 in 10 medical questions that had no safe answer. The rest of the time it answered anyway, fluent, confident, and often wrong.

So researchers are building tools to teach AI when to stay quiet. And the behavior they score as “safe,” again and again, is the same: when in doubt, the model should stop and say, “Please consult a doctor.”

That sounds right.

It is also where the thinking stops.

The whole approach rests on one assumption nobody checks: that the doctor you are sent to will do better.

Subscribe now

Start with the machine. When an AI gives a confident answer it has no real basis for, we now call it confabulation — making up a plausible story to fill a gap in what it actually knows. The field takes this seriously. There are now careful tools that catch a model in the act of inventing, by sampling its answers and flagging when its confidence is hollow. Good work, and necessary.

But confabulation is not a computer glitch. It is what any mind does, silicon or human, when it is unsure and feels pressure to produce an answer.

Daniel Kahneman spent a career proving this about people. He showed that confidence is not a readout of accuracy. It is a feeling, produced by how good a story the mind can tell itself. He called it the illusion of validity: we feel sure when the pieces fit, even when the evidence is thin. And he named the engine, the mind builds its answer from whatever is in front of it and quietly ignores what is missing.

Read that again. It describes a language model. It also describes a tired clinician at 3 a.m.

The evidence on doctors is not gentle. In a landmark 2008 review, two patient-safety researchers argued that physicians badly underrate how often their diagnoses are wrong. When one of them asked rooms full of doctors whether they had made a diagnostic error in the past year, only about 1 in 100 said yes. A 2013 study put numbers on the gap. Physicians got about 55 out of 100 easy cases right — and fewer than 6 out of 100 hard ones. Their confidence barely moved: about 7.2 out of 10 on the easy cases, 6.4 on the hard ones. Accuracy fell off a cliff. Confidence took a gentle stroll downhill. On the hard cases they were wrong more than nine times in ten and stayed nearly as sure of themselves.

What they almost never said was “I don’t know.” That phrase is in remarkably few doctors’ repertoires. The culture of medicine treats uncertainty as something close to failure, and tolerance for not-knowing measurably drops over the course of medical school. We train it out of people.

So here is the asymmetry the AI field keeps walking past. We have built detectors to catch the machine bluffing. We have built nothing of the kind for the clinician who does the same thing, undetected, and then writes it in the chart. The machine that confabulates gets a benchmark and a leaderboard. The human it defers to gets the benefit of the doubt.

In obstetrics this matters twice over, because there are two patients. And there is a second blind spot in the research: it treats silence as the safe choice. At the bedside it often is not. The harm in my field comes far more from things not done, the warning sign not named, the worrying tracing not acted on, the escalation that came too late, than from saying too much. An AI trained to go quiet whenever it is unsure can be trained straight into the passivity that hurts patients. And “consult a doctor” assumes there is a doctor to consult, and that the doctor knows. For many people typing questions at midnight, neither is true.

This is the part the silence debate misses, and it is the heart of a clinical opinion my colleagues and I just published. The danger was never that clinicians would use AI. The danger is that they will use it poorly, copying confident output without checking it, accepting an error because it reads well, treating a chatbot as an authority instead of a tool. The answer is not to avoid these tools, and it is not to make them quieter. It is competence: knowing when AI helps and when it does not, how to test its answers, how to catch a made-up drug dose before it reaches a patient, how to say plainly what is known and what is not, and who stays accountable when the screen goes dark. The clinician does.

Obstetrics has done this before. Ultrasound, electronic fetal monitoring, cell-free DNA screening, each arrived to resistance, and none of them replaced the obstetrician’s judgment. Each demanded a higher version of it. AI is next in that line.

But notice what that competence really asks for. Knowing the limits of your own knowledge. Saying “I don’t know” out loud. Matching your confidence to what you actually checked. That is exactly the calibration we are now trying to engineer into the machine — and it is the same skill medicine has never seriously required of itself.

For a patient, the practical version is small and powerful. The AI article you read before your appointment can sound certain and be wrong. So can the confident answer across the desk. The clinician worth trusting is not the one who never hesitates.

Share

It is the one who can say, “I’m not sure: let me check,” and then actually checks.

You are allowed to ask two questions out loud: How sure are you? And did you verify it? A good doctor will not flinch at either.

My take: we are studying the wrong half of the problem. The question was never silence versus speech. It is calibration — knowing what you know, and admitting what you don’t. Kahneman built his entire method on the premise that he could not trust his own confidence, so he backed everything with data. That is the discipline. It is rare in the model. It is just as rare in the physician we hold up as the safe alternative.

The arrival of AI is forcing a calibration test on medicine that medicine never volunteered for. We should take it. The goal is not a quieter machine. It is a profession — human and artificial together — that is finally honest about the edge of its own knowledge. In obstetrics, where two patients depend on that honesty, “first do no change” has to give way to “first do good.” Competence, not silence, is the safer answer.

Bottom line: don’t ask how to make AI stay quiet. Ask how to make both the machine and the clinician tell you the truth about how much they actually know.

If this is the kind of thinking you want more of, subscribe to ObGyn Intelligence. It is free, it is independent, and it says plainly what the data shows.

References

1. Presacan O, Nik A, Ojha J, Thambawita V, Ionescu B, Riegler MA. When silence is safer: a review and decision-theoretic framework for LLM abstention in healthcare. npj Digit Med. 2026. doi:10.1038/s41746-026-02882-1.

2. Grünebaum A, Dudenhausen J, Chervenak FA. Clinical artificial intelligence competence in obstetrics and gynecology: patient safety, physician accountability, and responsible use. Am J Obstet Gynecol. 2026. doi:10.1016/j.ajog.2026.06.011.

3. Berner ES, Graber ML. Overconfidence as a cause of diagnostic error in medicine. Am J Med. 2008;121(5 Suppl):S2-S23. doi:10.1016/j.amjmed.2008.01.001.

4. Meyer AND, Payne VL, Meeks DW, Rao R, Singh H. Physicians’ diagnostic accuracy, confidence, and resource requests: a vignette study. JAMA Intern Med. 2013;173(21):1952-1958. doi:10.1001/jamainternmed.2013.10081.

5. Kahneman D. Thinking, Fast and Slow. New York, NY: Farrar, Straus and Giroux; 2011.

6. Grünebaum A, Chervenak J, Pollet SL, Katz A, Chervenak FA. The exciting potential for ChatGPT in obstetrics and gynecology. Am J Obstet Gynecol. 2023;228:696-705. doi:10.1016/j.ajog.2023.03.009.

You no longer need an engineer to turn what you know into something patients can use.

“Tools will come and tools will go. Only the vibe coder remains.” A legendary music producer said it about software. It is also the best argument I know for why doctors should learn to build their own clinical tools. Here is what it means for medicine.

Andrej Karpathy, a founder of OpenAI, gave it a name in February 2025. Rick Rubin, who has produced more hit records than almost anyone alive, gave it a philosophy: “Tools will come and tools will go. Only the vibe coder remains.” Collins Dictionary named it the Word of the Year for 2025. Vibe coding has arrived.

Subscribe now

Most doctors are watching from the sidelines. They should not be.

What vibe coding actually is

Vibe coding means building software by describing what you want in plain language and letting an AI write the code. You do not memorize syntax. You do not hire an engineer. You say what the tool should do, the machine builds it, you test it, and you correct it. Karpathy’s original line was to forget that the code even exists. You see something, say something, run it, fix it.

Rubin’s point cuts deeper. He produces records without playing the instruments. His value is taste and judgment, knowing what good sounds like and steering the artist toward it. The tool in front of him keeps changing. His ear does not. That is the part that lasts.

For a doctor, the ear is clinical judgment

Fifty years at the bedside, thousands of patients, the feel for what a number means and when it matters. That is the scarce thing. Until now, turning that judgment into a working tool required someone else. You explained the idea to a programmer who did not know obstetrics, waited months, paid a great deal, and often got back something that missed the clinical point. The translation step was the bottleneck. Vibe coding removes it.

My own numbers

I have built more than 100 clinical tools this way. Risk calculators, screeners, patient education pages, decision aids, all on tools.obmd.com and liveevidence.com, all created by directing AI rather than writing code by hand. To my knowledge I am the only ObGyn in the United States with a tool library of this size built this way. I am in my seventies. I do not say this to impress you. I say it because if I can do it, the resident who grew up with a phone in her hand certainly can.

The tools are the easy part. What doctors get wrong about vibe coding is the part that actually matters.

The judgment is the product

A software engineer can build you a calculator. He cannot tell you whether the threshold is right, whether the population it was validated in looks like your patient, or whether the output will frighten a woman at two in the morning over a number that does not warrant it. You can. The clinical judgment is not a feature you add at the end. It is the whole point.

This is why doctors, not engineers, should hold the controls. The tool is only as good as the question behind it, and the question is clinical. When I build a tool, I decide what counts as a meaningful number. I write the plain-language explanation. I choose what the tool refuses to say because the evidence does not support saying it. Those are not coding decisions. They are medical ones.

Where vibe coding gets dangerous

Now the honest part. Karpathy’s advice to accept everything and forget the code exists is fine for a personal weekend project. It is reckless for a tool a patient will trust. A tool that computes a risk score is making a medical claim. It deserves the same scrutiny as a drug or a device.

So the rules are simple. Every number in the tool traces to a published source. I check the math against the original model, by hand, before anything goes live. I do not ship a black box: if a tool gives a result, it shows its reasoning and cites its evidence. And no patient data is ever stored or transmitted. My tools run entirely in the browser and keep nothing. Privacy is not a setting you switch on. It is the design.

Here is the paradox worth sitting with. The thing that makes a doctor safe at vibe coding is the same thing that makes vibe coding dangerous in untrained hands: judgment about what the numbers mean. A layperson who vibe codes a pregnancy risk tool does not know the threshold is wrong. You do. That is not a small advantage. It is the entire case for why this skill belongs to clinicians.

What this means for patients

When your doctor uses a tool to estimate your risk, you have a right to ask three things. Where do the numbers come from. Has the tool been checked against the published evidence. And what happens to the information you typed in. A good tool answers all three plainly. A tool built by a clinician who understands the evidence can answer them. A black box bought from a company often cannot.

The skill outlasts the tool

The tool I use today will be gone in two years. Karpathy himself, one year after coining the term, already calls vibe coding passe and renames the professional version agentic engineering. The label keeps changing. The brand of AI keeps changing. None of that matters. What Rubin understood, and what doctors should understand, is that the durable skill is not the tool. It is knowing what good looks like and being able to direct the machine toward it. For a doctor, that is simply clinical judgment pointed at a new instrument.

Conclusion

I am not telling you to become a programmer. I am telling you the opposite.

The era of the doctor as the person who memorizes ten thousand facts is ending, because the machine remembers better than any of us.

What it cannot do is care for the specific patient in front of you, or decide what a number should mean for her.

That is yours.

Vibe coding lets you build the tools that judgment deserves, without waiting for permission, a budget, or a programmer who has never set foot in a delivery room. Rubin reimagined the Tao Te Ching for the age of code. I will borrow from a source he would appreciate, Marcus Aurelius: the obstacle is the way. For doctors, the tool was never the obstacle. The translation was. It is gone. Build the thing only you can build.

ObGyn Intelligence is independent because readers keep it that way. If this changed how you think about the next decade of your work, subscribe. The evidence, and now the tools, should travel.

Subscribe now

References

1. Rubin R, Andreessen M, Horowitz B, Midha A, Torenberg E. Rick Rubin: Vibe Coding is the Punk Rock of Software. a16z Podcast. 2025. Available from: https://a16z.com/podcast/rick-rubin-vibe-coding-is-the-punk-rock-of-software/

2. Karpathy A. Post on X introducing the term “vibe coding.” February 2, 2025.

3. Collins Dictionary. Word of the Year 2025: vibe coding. 2025.

4. Taft DK. Vibe coding is passe. Karpathy has a new name for the future of software. The New Stack. February 10, 2026. Available from: https://thenewstack.io/vibe-coding-is-passe/

ObGyn Intelligence - Evidence Matters | obmd.com

A patient comes back for her second visit. Her symptoms have shifted. Her lab results are in. A good doctor remembers what happened last time, reads the new numbers, and adjusts the plan. That thinking over time is the hard part of medicine. The one-shot diagnosis is the easy part.

Subscribe now

A new study in Nature asked a simple question: can an artificial intelligence do the hard part?

AMIE stands for Articulate Medical Intelligence Explorer.

It is a medical AI system built by Google, running on their Gemini model. Earlier work showed it could take a patient history and suggest a diagnosis. This new paper pushes further, into what doctors call management — deciding what tests to order, what to prescribe, what to watch, and how to change course as a patient returns over weeks.

The design is clever, and worth understanding in plain terms. AMIE is built from two parts that work like the two speeds of human thinking.

• One part is a fast talker. It chats with the patient, sounds warm, and keeps track of the conversation.

• The other part is a slow thinker.

While the conversation continues, it quietly reads through hundreds of pages of clinical guidelines and writes a careful plan, with every recommendation tied to the exact guideline it came from. The idea comes straight from Daniel Kahneman’s Thinking, Fast and Slow: quick intuition on the surface, slow reasoning underneath.

To test it, the researchers ran a blinded, randomized exam. They wrote 100 made-up cases, each unfolding over three visits a day or two apart. The cases spanned five specialties, and obstetrics and gynecology made up 20 of the 100. Trained actors played the patients. For each case, the same patient was seen once by AMIE and once by one of 21 primary care doctors, all by text chat. Specialists then graded both, without knowing which one was the machine.

AMIE did at least as well as the doctors on overall management, and clearly better on two things. First, it was more precise. Instead of “prescribe an antibiotic,” it named the drug, the dose, the route, and the duration. On treatment precision it scored around 95 out of 100; the doctors scored in the 60s. Second, it stuck closer to the guidelines and showed its work, citing the source for each recommendation. When graders did prefer one over the other, they picked AMIE far more often than the doctor — though in about half the cases they saw no real difference at all.

Here is the part the headlines will skip. The doctors were set up to lose. They came from Canada and India but were judged against United Kingdom guidelines they do not use every day. They worked through a text box, not the phone or video they normally use. The cases had clean, correct answers, which real patients rarely provide. And the visits were one or two days apart, not the weeks the cases described — so human memory was tested in a way that does not match real care. AMIE, meanwhile, was built and tuned for exactly this setup. This is what I call the missing comparator. When you compare a system to people working with one hand tied, “the AI won” is not the finding. The finding is narrower, and still useful: the AI is very good at reading guidelines and writing precise plans.

Subscribe now

Why should an ObGyn care? One in five of these cases was ours. The things AMIE did best — remembering the last visit, citing the right guideline, writing a precise plan — are exactly the things that get sloppy on a busy, tired labor floor. A tool that never forgets and always cites the source has real value there. But the things AMIE cannot do are also ours: the woman in front of you whose values, fears, and body do not match the guideline. The study’s own authors say plainly that AMIE is not ready for real patients. They are right.

My take. This is careful, honest work, and I do not say that lightly about an AI paper. The two-speed design is smart, grounding the plan in guidelines is the right instinct, and the authors were more open about their limits than most. But “AI matches doctors” is not what this shows. It shows that AI beats a distracted human at the one thing computers have always been better at: holding everything in memory and quoting the manual exactly. That is useful. It is not judgment. For ObGyn the lesson is neither fear nor worship. Let the machine carry the memory and the guideline lookup, and keep the thinking about the specific woman where it belongs — with you. The real risk is not that AMIE replaces us. It is that we stop practicing the parts it does well and forget how to do them ourselves.

Share

One footnote, for the attentive reader. The study was funded by Google, and the authors may own Google stock. I mention this not as an accusation but as arithmetic.

AMIE is real, it is impressive, and it is a research demonstration, not a doctor. Read the result, not the headline.

References

1. Liévin V, Palepu A, Weng W-H, Saab K, Stutz D, Cheng Y, et al. Towards Conversational AI for Disease Management. Nature. 2026. doi:10.1038/s41586-026-10764-5.

2. Tu T, et al. Towards conversational diagnostic artificial intelligence. Nature. 2025;642:442–450.

Last week I went looking for one old document in ChatGPT. I found something I did not expect. There was a folder holding every file I had ever uploaded. Old drafts. PDFs. Documents I was sure I had deleted months ago. I had deleted the conversations. The files were still sitting there.

The folder is called the Library.

Most people do not know it exists.

It’s in your left sidebar.

In front of your eyes.

That is the whole problem in one sentence.

Here is what the Library is.

When you upload a file to ChatGPT, or when ChatGPT makes a file for you, it saves a copy. Documents, spreadsheets, pictures, PDFs, all of it. You do not turn this on. You do not click save. It happens on its own, quietly, in the background. The files pile up whether you notice or not.

Now here is the part that matters most. Deleting the chat does not delete the file. When you delete a conversation, it disappears from the list on the side of your screen. It looks gone. But the file you uploaded stays in the Library. So the thing almost all of us believe, that deleting a chat means it is gone, is simply not true. The chat is gone. The file is not.

Think about what people put into ChatGPT. A resume. A lease. A letter from a lawyer. A photo of a bill. A medical record. For my colleagues in medicine, think about every case we called de-identified that was not quite as anonymous as we told ourselves. Any of it can be sitting in storage you never knew you had, on a company’s computer, not yours.

• Opt-in means a person’s data is not used unless they actively give permission.

• Opt-out means the data may be used by default unless the person notices the setting and turns it off.

Opt-in is safer because it respects informed consent, reduces accidental data sharing, and places the burden on the organization to obtain permission before using personal information. Opt-out systems are weaker because many people never see the setting, do not understand it, or assume privacy is already protected.

Subscribe now

There is a second layer, and it is also turned on by default. ChatGPT has a setting called Improve the model for everyone. When it is on, the company can use what you type, and the files you upload, to train its future models. On by default. Most people never see the switch.

And when you finally do delete a file the right way, the company removes it from its servers within thirty days. So delete does not mean gone today. It means gone in about a month. That is better than forever. It is also not nothing.

So what do you do about it? Three things, and none of them take long.

1. First, turn off training. Click your picture in the top right corner of the screen. Open Settings, then Data Controls. Find the line that says “Improve the model for everyone” and switch it OFF. From that moment on, your new chats are not fed into the next model.

   

2. Second, clean out the Library in one go. Open the Library from the menu. At the top of the list of files you will see the word Name. Click it. A box appears that lets you select everything at once. Check it, pick what you do not want kept, and delete. You do not have to click each file one at a time, which matters when there are hundreds of them.

   Subscribe now

   

   

3. Third, use Temporary Chat for anything private. The button sits in the upper right of the chat window. Looks like a broken circle. Not easy to find. A Temporary Chat is never saved to your Library and is never used for training. It is the right tool for the one-off question, the symptom you are embarrassed about, the document you do not want kept, the thing you would not say with the door open.
   Here it is, upper right of a new chat:

What does this mean for you?

Not that ChatGPT is dangerous (or maybe it is?).

Share

Saving your files so you can find them later is a fair and useful feature. For some.

The problem is not the Library. The problem is the silence around it. The saving is on by default. The deleting does not work the way the screen makes you think it does. And almost no one is ever told. That is not a technology problem. That is a consent problem.

Here is my take. I have spent fifty years telling patients the same thing. Consent only counts when it is informed. A choice you were never told about is not a choice you made. It is something that was decided for you while you were looking somewhere else. The rule that governs a procedure should govern your data too. You have a right to know what is being kept, what is being used, and how to stop it.

Share

So open your Library today. Look at what is in there. You may be surprised, the way I was. Then decide, with your eyes open, what stays and what goes.

It took me one click to find the folder. It took one more to empty it. The second click is the one worth telling people about.

If this saved you from leaving something private on a stranger’s server, it did its job. ObGyn Intelligence is where I write about exactly this, the gap between what we are told and what is true, in medicine and in the tools we now use every day. Subscribe, and bring the friend who still thinks deleting the chat deletes the file.

Sources

OpenAI. File storage and Library in ChatGPT. OpenAI Help Center. Available from: https://help.openai.com/en/articles/20001052-file-storage-and-library-in-chatgpt

OpenAI. What if I want to keep my history on but disable model training? OpenAI Help Center. Available from: https://help.openai.com/en/articles/8983130

Two pathologists look at the same breast biopsy. One sees cancer or another anomaly. One does not. This is not a thought experiment. In a 2015 study in JAMA, 115 pathologists read the same slides. Across the full spectrum of breast findings — benign, atypia, DCIS, invasive — overall agreement with the expert panel was 75 percent, with the disagreement concentrated in the preinvasive gray zone.” The cases that mattered most — the early, borderline changes — were the ones they disagreed on most. Same slide. Same patient. A different answer depending on who held the microscope.

I keep that number in mind whenever someone tells me medical AI cannot be trusted.

A new editorial in the International Journal of Medical Informatics makes a strong claim right in its title: almost all machine learning models for medicine are wrong. The authors are serious people, and part of their argument is right.

The labels used to train these tools come from human experts who disagree.

The models are tested on the same kind of patients they learned from, not on new hospitals or new populations. They are scored with numbers that look exact but shift the moment the setting changes. And once switched on, they are rarely watched, even though medicine keeps changing underneath them.

Subscribe now

The title is not new.

It echoes a famous 2005 essay, “Why most published research findings are false.” That one was about human research — the same studies and expert opinions these models are trained to copy. Follow its logic and almost all of medicine rests on shaky findings too. Somehow the cannon gets pointed only at the machine.

Share

All true, as far as it goes. I have made versions of these points myself. Build a tool on a weak foundation, never test it in the real world, and you should not trust it. But the paper never asks the one question that matters most. Wrong compared to what?

Every flaw on their list is also a flaw in human doctors — usually a bigger one. Doctors disagree with each other, as those pathologists did.

They are trained once, tested once, and then practice for decades. Nobody runs a doctor through a second hospital to confirm she still performs. Nobody measures, year after year, whether her judgment still matches the evidence. And the knowledge she learned in training is often more than a decade behind the best current science — on average it takes about 17 years for research to reach everyday practice. A machine can be corrected overnight. A habit takes a generation.

So when a study calls a model “wrong,” I want to know: wrong next to what human standard?

The human standard is not perfect truth.

It is a group of people who disagree, carry old knowledge, and are almost never checked.

This is where the paper trips over its own feet. Its strongest section explains that there is no clean truth in medicine. Diagnoses are fuzzy. Lab tests vary. Experts split. The authors say so plainly. Then they turn around and call the machine “wrong” — wrong against the same fuzzy standard they just told us not to trust.

You cannot have it both ways. If the ruler is made of rubber, the word “wrong” loses its meaning. When a model disagrees with a label, it often disagrees with a label a second expert would have rejected too. That is not error. That is the ordinary disagreement of medicine, renamed as a machine defect.

So who is the arbiter of truth? In medicine, there isn’t one. There is a committee of fallible humans, and we have agreed to treat their majority vote as the truth. That is fine — as long as we judge the machine by the same yardstick we use on ourselves, not a stricter one we invented the day the machine walked in.

Subscribe now

I will give the authors one real point: scale. One tired doctor harms patients one at a time. A bad model, switched on across a whole hospital system, can harm thousands before lunch. That is a genuine reason to watch deployed tools closely — to test them, keep them honest, and check them over time. Their list of fixes is a good list. I would sign most of it. But that is an argument about how we deploy these tools, not about whether they are “wrong.” The methods in this paper are right. The frame is a double standard wearing a white coat. We are demanding proof from the machine that we never demanded from ourselves.

My position is simple.

Hold medical AI to a high standard.

Then hold human medicine to the exact same one.

Stop asking whether the model is perfect.

Start asking whether it is better than what we already do — and then actually measure what we already do, which we almost never have. A tool does not need to be flawless. It needs to beat the disagreeing, aging, unchecked standard we already live with. Often, it already does.

The next time you read that almost all medical AI is wrong, ask the question the headline skips: compared to whom?

Share this with a colleague who still calls the human exam the gold standard — the data on how often we agree should travel.

ObGyn Intelligence is free because the work matters. If you want to keep it independent and in your inbox, subscribe.

References

1. Cabitza F, Jurman G, Molinari F, Bellazzi R. Why almost all ML models for medicine are wrong — and what we need for evidence-based medical AI. Int J Med Inform. 2026 (in press). doi:10.1016/j.ijmedinf.2026.106538.

2. Elmore JG, Longton GM, Carney PA, et al. Diagnostic concordance among pathologists interpreting breast biopsy specimens. JAMA. 2015;313(11):1122-1132. doi:10.1001/jama.2015.1405. PMID 25781441.

3. Morris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011;104(12):510-520. doi:10.1258/jrsm.2011.110180. PMID 22179294.

4. Ioannidis JPA. Why most published research findings are false. PLoS Med. 2005;2(8):e124. doi:10.1371/journal.pmed.0020124. PMID 16060722.

Ask any AI chatbot a medical question and you get an answer in seconds. It sounds calm, organized, and sure of itself. That last part is the problem. Confidence is not the same as reliability, and a chatbot has no way to tell you which of its sentences it would stake its life on and which it is simply guessing.

So let me show you something simple. I asked a medical AI one ordinary clinical question — the kind a pregnant woman or a first-year resident might ask. Then I asked it the same question a second time, word for word, with a single instruction added to the end. Same model. Same minute. Two very different answers.

Here is the first one.

Prompt: “Tell me what blood tests a pregnant woman should do at her first prenatal visit.”

Here is the answer:

“At her first prenatal visit, a pregnant woman should have her blood type and Rh factor checked, along with an antibody screen and a complete blood count to look for anemia. She should be tested for rubella immunity, hepatitis B, hepatitis C, HIV, and syphilis, and screened for chlamydia and gonorrhea. Her thyroid should be checked with a TSH test, and she should be offered genetic carrier screening and screening for Down syndrome. A urine test is also part of the standard first visit.”

Not wrong, exactly. A reasonable list. But notice what it does not do. It never separates the tests every pregnant woman gets from the ones that depend on her age, weight, or history. It presents a thyroid test as routine — when the main US guideline actually recommends thyroid testing only for women with specific risk factors. It cites nothing. And it never once tells you what it would need to know about the patient to answer well. It just talks.

Now the same question, with one instruction added to the end.

That instruction is the Clinical Evidence Prompt or the CEPrompt™.

PROMPT: “(Tell me what blood tests a pregnant woman should do at her first prenatal visit [PLUS THE INSTRUCTION, the CEPrompt™.”

Here is the full answer it produced — nothing trimmed.

“This is the US/ACOG framework at a routine first visit, assuming no results already documented this pregnancy.

Universal — standard of care for all (high certainty): blood type, Rh(D) and antibody screen; CBC; rubella IgG; hepatitis B (HBsAg minimum, triple panel if not previously documented); hepatitis C (every pregnancy); HIV (opt-out); syphilis (repeat in third trimester and at delivery). Urinalysis and urine culture are also routine but not blood-based.

Selective — depends on the individual patient (moderate certainty): chlamydia/gonorrhea (universal if under 25, otherwise risk-based); varicella IgG if no immunity history; hemoglobinopathy evaluation guided by CBC/MCV and ancestry; early diabetes screening if overweight/obese with risk factors or prior GDM; lead and TB testing if at risk.

Offered, not assigned: genetic carrier screening and aneuploidy screening (cell-free DNA or serum) — both offered to all regardless of age.

Evidence vs. position: The universal panel reflects guideline consensus. Thyroid (TSH) is the exception — ACOG (PB 223) recommends selective, risk-based testing, not universal. Presenting routine TSH as standard is an inference, not consensus; universal offering is a defensible minority position, not the guideline.

Missing information that changes the selective list: maternal age, BMI, ancestry, prior obstetric/GDM history, immunization and prior-testing records, STI risk, gestational age at presentation.

Certainty: high for the universal panel; patient-dependent for selective tests; the thyroid point is genuinely contested.

Citations: Viral Hepatitis in Pregnancy. Clinical Practice Guideline No. 6. ACOG. Obstet Gynecol 2023;142:745–59. SMFM Consult Series #69: Hepatitis B in pregnancy. Am J Obstet Gynecol 2024;230:B2–B11. Owens DK, et al. USPSTF hepatitis C screening. JAMA 2020;323:970–5. doi:10.1001/jama.2020.1123.”

Read that again, because every line of it is doing a job the first answer skipped. It splits the tests into three groups instead of one: the ones every pregnant woman gets, the ones that depend on her age, weight, ancestry, or history, and the ones that are offered for her to accept or decline. It draws a clear line between guideline consensus and the author’s own position — and it catches the thyroid trap, telling you plainly that routine TSH for everyone is not what the main US guideline recommends, even though the first answer presented it as routine. It tells you what it would need to know about the actual patient before the selective list can be finalized. It labels how certain it is, group by group. And it ends with real citations you can open and check, not a vague wave at “the guidelines.”

Subscribe now

That is the difference between a list and a consultation. The first answer hands you tests. The second one reasons with you — it shows what is settled, what depends on the patient, what is contested, what is missing, and where it got each claim. A clinician can act on the second answer. A patient can ask better questions because of it. Nobody should act on the first.

Same machine. The model did not get smarter between the two answers. The second answer is better because the prompt told it to show its work — to separate what the evidence supports from what it is assuming, to say how sure it is, to name what it is missing, and to admit when something is genuinely debated.

That instruction is the whole trick.

That instruction is the Clinical Evidence Prompt or the CEPrompt™.

It is one sentence.

You paste it at the end of any clinical question, and it changes the kind of answer you get back. Not the topic. The honesty.

The exact wording of the CEPrompt™ — the sentence that turns the first answer into the second — is below, with the six things it does and how to attach it to any question you ask. Paid subscribers read on.

Read more

By 2026, about one in three American adults had asked an AI chatbot for health advice. Among physicians, 81% report using artificial intelligence in their work. And fewer than 15% have had any formal training in how to use it safely. Read those three numbers again. We have adopted a clinical tool faster than almost any technology before it, and we skipped the part where someone teaches us how to use it.

A large language model — ChatGPT, Claude, Gemini — is not a database of checked facts.

It is a very good pattern engine that predicts the next likely words.

Most of the time that is helpful. Sometimes it invents a citation that does not exist, or a dose that is wrong, and it does so with total confidence. The danger is not that the tool is bad. The danger is that it is convincing, and that we are using it without knowing where it breaks.

It can make mistakes. It hallucinates. But with proper training you can get the most out of it.

So we built the course.

It is called AI Health Course. Passwords are required to get in. The course is in private preview. Paid subscribers get the access password below after the paywall.

Please message me on LinkedIn for any questions.

It is short — about three to four hours, in sittings of fifteen to thirty minutes — and it is written both for patients and clinicians, not engineers. It covers how to give the model the right context, how to spot an answer that is unsafe, how to verify what matters, how to keep patient information private, and how to stay responsible for the decision. Every idea is taught in plain language, then practiced on a real, de-identified example.

Here is the part I like. I built the course itself by “vibe coding.” That means I described what I wanted in plain English, Claude wrote the working code, and I tested and corrected it until it was right, no web developer in the room. But nothing was taken on the model’s word. Every clinical number, every citation, and every privacy rule was checked against the original source before it went in. The tool drafts; the clinician stays accountable. That is the whole message of the course, and it is also how the course was made.

The course is in private preview.

Paid subscribers get the access passwords below.

Subscribe now

Read more

The Society for Maternal-Fetal Medicine (SMFM) regularly posts an “Ultrasound Case of the Week Challenge.” It’s meant to sharpen our diagnostic instincts and celebrate the art of pattern recognition.

But the exercise now has a rival: artificial intelligence that can see what we see—and explain it better.

Subscribe now

The Case (as posted by the Society of Maternal-Fetal Medicine - SMFM)

Case Summary:
A patient is referred to Maternal-Fetal Medicine at 32 weeks’ gestation because of ultrasound findings of short long bones. On detailed scan, you confirm:

• Shortened limbs with mild femoral bowing

• Normal ossification

• No evidence of fractures

• Amniotic fluid index (AFI): 29 cm (polyhydramnios)

• No other structural anomalies noted

Question: Based on these findings, what is the most likely diagnosis?

Options:
A. Thanatophoric dysplasia
B. Achondroplasia
C. Osteogenesis imperfecta
D. Campomelic dysplasia
E. Diastrophic dysplasia

Correct Answer: B. Achondroplasia

The AI Prompt

Prompt to ChatGPT-5:
“You are a Maternal-Fetal Medicine specialist. Interpret this 32-week fetal ultrasound showing short long bones, mild femoral bowing, normal ossification, no fractures, and an AFI of 29 cm. What is the most likely diagnosis? Explain your reasoning, discuss the differential, outline recommended management, and address ethical issues related to prenatal counseling.”

ChatGPT’s Full Response (abridged for clarity)

Diagnosis:
The findings are most consistent with achondroplasia, the most common non-lethal skeletal dysplasia.

Reasoning:

1. Limb pattern: Shortening of long bones (rhizomelic pattern—proximal > distal) with mild bowing is typical.

2. Ossification: Normal echogenicity distinguishes achondroplasia from lethal dysplasias like thanatophoric dysplasia, which show poor ossification or platyspondyly.

3. Fractures: Absence of fractures argues against osteogenesis imperfecta.

4. Thorax: Normal size (not narrow) suggests adequate pulmonary development, differentiating from lethal dysplasias.

5. Polyhydramnios: Common secondary feature due to limited fetal swallowing or altered skeletal mechanics.

Differential Diagnosis:

• Thanatophoric dysplasia: More severe bowing, narrow chest, poor ossification.

• Campomelic dysplasia: Anterior bowing of tibia/fibula, ambiguous genitalia possible.

• Diastrophic dysplasia: Hitchhiker thumb, clubfoot, cystic ear swelling.

• Osteogenesis imperfecta: Multiple fractures, hypomineralization.

Management Plan:

1. Confirmatory Testing:

   • Offer FGFR3 genetic testing (dominant mutation) by amniocentesis if diagnosis uncertain.

   • Counsel parents about inheritance (often de novo, but 50% recurrence if one parent affected).

2. Monitoring:

   • Serial ultrasounds for growth, AFI, and head circumference (often macrocephalic).

   • Assess fetal well-being and plan delivery at a tertiary center with neonatal support.

3. Delivery Planning:

   • Mode of delivery: Vaginal delivery often possible, but cephalopelvic disproportion may occur; cesarean may be required if fetal head size exceeds maternal pelvic dimensions.

4. Postnatal Care:

   • Neonatal evaluation to confirm diagnosis.

   • Early referral to genetics and orthopedics.

Ethical Considerations:

• Autonomy: Parents must be given accurate, balanced information—achondroplasia is non-lethal and compatible with a normal lifespan.

• Non-maleficence: Avoid framing as “abnormal” or “lethal.” Words matter.

• Justice: Families deserve equal access to expert counseling regardless of geography; AI-assisted image interpretation could bridge that gap.

• Beneficence: Encourage connection with patient support networks (e.g., Little People of America).

Summary:

Most likely diagnosis: Achondroplasia.
Rationale: Rhizomelic limb shortening, normal ossification, no fractures, normal chest, polyhydramnios.
Next steps: Confirm via FGFR3 testing, multidisciplinary perinatal care, delivery planning, and ethical counseling.

The Meta-Lesson

In this single exchange, the AI performed the entire intellectual arc of a case challenge: observation, reasoning, differential, testing, management, and ethics. It didn’t just choose the right answer—it articulated the why and the how.

What SMFM’s quiz presents as a game of recall, AI transforms into a masterclass in reasoning. The educational format can evolve: the point isn’t whether the AI knows, but whether we can understand what it knows.

Imagine if each SMFM “Case of the Week” included a “ChatGPT-5 analysis” sidebar. Learners could study the logic, question assumptions, and compare human versus machine reasoning. Instead of multiple choice, the challenge becomes interpretive: Do you agree with the AI’s rationale? Why or why not?

That is true diagnostic education—understanding, not guessing.

The Ethical Frontier

Is it ethical to use AI in prenatal diagnosis?

Today, the stronger argument is that it is unethical not to use AI.

When a technology demonstrably enhances diagnostic accuracy, speeds interpretation, reduces disparities, and helps explain findings in language families can understand, deliberately avoiding it violates the physician’s duty to beneficence and nonmaleficence.

The core ethical obligation of medicine is to use the best available means to prevent harm. A clinician who ignores AI-assisted tools, despite knowing they outperform unaided human interpretation in certain pattern-recognition tasks, is choosing a lower standard of care. This is not professional humility; it is ethical inertia. If AI can flag skeletal dysplasia earlier, identify anomalies that could otherwise be missed, or prepare families for counseling with greater clarity, then refusing to integrate it is no different from refusing to use an improved diagnostic test because one prefers the old microscope.

AI is not a replacement for the physician’s judgment—it is a moral assistant that strengthens it. Used responsibly, it widens access to expertise, supports overworked clinicians, and makes high-quality prenatal imaging interpretation available in underserved regions where MFM specialists are scarce. In that sense, AI becomes an equalizer, not an intruder. To withhold such a tool from patients because of habit, pride, or fear of change is to violate the principle of justice.

Of course, ethical use demands oversight. Clinicians must understand when the model’s confidence is justified, when to override it, and how to explain uncertainty transparently. Accountability remains human. But to reject AI altogether when it can reduce diagnostic errors, improve counseling, and spare families unnecessary suffering would be a failure of professional duty.

Subscribe now

The true ethical risk today lies not in using AI, but in refusing to—thereby accepting preventable errors, slower recognition, and less informed care. History has seen this pattern before: skepticism toward antisepsis, ultrasound, and even fetal monitoring delayed their adoption and cost lives. In the same way, ignoring validated AI assistance in obstetric imaging is no longer a neutral choice. It is an ethically indefensible one.

There are many precedents where not using available technology became ethically unacceptable once its benefits were proven.

Refusing to perform ultrasound in pregnancy, when it can detect life-threatening conditions like placenta previa or anencephaly, would be indefensible.

Declining electronic fetal monitoring during labor in high-risk cases would ignore decades of evidence linking timely intervention to improved neonatal outcomes.

Omitting Rho(D) immune globulin after an Rh-negative pregnancy, or refusing to use antibiotic prophylaxis for cesarean delivery, would likewise violate the duty to prevent avoidable harm. In the same way, neglecting validated AI diagnostic assistance when it can enhance accuracy and patient understanding now falls within that same ethical category—a failure to use the best tools medicine has to protect mothers and babies.

Share

Reflection / Closing

The “Ultrasound Case of the Week” once tested what we could see. Soon, it may test how we think alongside AI. The ultimate goal is not to outguess the machine but to outgrow our dependence on guessing. The next generation of maternal-fetal medicine will belong to those who can merge evidence, empathy, and ethics—whether the first diagnosis comes from a human hand or a silicon brain.

What vibe coding is, why it matters, and why an ObGyn professor is further along than most health systems.

A resident came to me last year with a clinical question she had already run through a large language model. The answer was good. Her instinct to use AI was correct. What surprised me was not her question but her hesitation to admit where she had looked first.

She was using the technology but treating it like a secret.

I understand the hesitation. Medicine moves slowly, and the slowest part moves fastest right now. Something has changed in the last year that most clinicians have not yet named. The name is vibe coding.

What Vibe Coding Is

The rest of this post is for paid subscribers.

Read more

If I can see the file, it exists.

If it is in iCloud, it is protected.

If it appears on more than one computer, it must be backed up.

That is what I thought.

After more than 40 years of using Apple computers, my MacBook crashed. Only then did I discover that iCloud had silently stopped updating some of my files for about a month. I had assumed that because my work was in iCloud, it was safe. I saw them in my iCloud folders. I saved them there. All my folders were in iCloud. Or so I thought.

They were not.

The problem was not that I had ignored computers. The problem was that the system uses familiar words in confusing ways.

It’s what I teach about patient safety. Rarely the individual, mostly the system.

Most people hear “sync” and think “backup.” That is understandable. If a file moves from a laptop to iCloud and then appears on another laptop, it feels protected. It feels as if there is another safe copy. Big Mistake!

But the real test is not what the file looks like on a normal day.

The real test is what happens when something goes wrong.

If I accidentally delete the file, is there a separate recovery copy?

If I overwrite the file, can I get the older version back?

If my laptop stops sending files to iCloud, will I know? (Hint: You won’t necessarily know)

If a file appears in Finder but is not physically downloaded to my computer, will my backup service copy it?

This is where the difference becomes painfully real.

iCloud is excellent at making files available across devices. That is its main job. A true backup has a different job: it keeps a separate recoverable copy when the current file is lost, damaged, deleted, overwritten, or never uploaded.

A simple way to say it is this:

iCloud helps your files appear in more places.

A backup helps your files survive when something goes wrong.

That difference should be obvious. It is not. Apple does not make it obvious. The icons are small. The settings are confusing. The word sync sounds reassuring. The word cloud sounds even more reassuring. But reassurance is not the same as recovery.

For physicians, professors, researchers, editors, writers, and students, this matters. Our files are not just files. They are manuscripts, lectures, peer reviews, grant drafts, response letters, legal documents, patient safety projects, and years of intellectual work.

This article is for people like me: educated, busy, careful, and still surprised to learn that seeing a file, storing a file in iCloud, and having a true backup are not the same thing. Trust me, I know it. Following the explanations and instructions below will save you many times the price you pay for a yearly ObGyn Intelligence subscription.

I paid many more times the price.

Read more

That is not because we are careless. It is because the system is designed to make us stop thinking about it.

We open a laptop. We see a folder. We see a file. The same file may appear on another computer. It may be visible from a phone. It may be in OneDrive, Google Drive, Dropbox, Box, or another cloud service. The natural conclusion is simple:

The file is safe.

That conclusion may be wrong.

This is not only an Apple or iCloud problem. The same confusion exists on Windows computers and in nearly every cloud storage system. The names change, but the risk remains the same.

On a Mac, the file appears in Finder.

On a Windows computer, the file appears in File Explorer.

In both cases, the file may look like it is on the computer. But sometimes it is not fully stored there. It may be mostly in the cloud. It may be online-only. It may be waiting to download. It may be visible as a placeholder. It may be available only if the internet connection works.

Subscribe now

That distinction is easy to miss.

Most people hear the word “sync” and think “backup.” That is understandable. If a file moves from my computer to the cloud and then appears on another computer, it feels backed up. It feels as if there is now another safe copy.

But the real test is not what happens on a normal day.

The real test is what happens on a bad day.

• If I accidentally delete the file, is there a separate recovery copy?

• If I overwrite the file, can I recover the older version?

• If my computer stops uploading files to the cloud, will I know?

• If a file appears in File Explorer but is not physically downloaded, will my backup service copy it?

• If the cloud service changes a file everywhere, do I still have an independent copy from before the change?

This is where the difference becomes real.

Cloud services are excellent at making files available across devices. That is their main job. They let us work from a desktop, laptop, tablet, phone, office computer, or browser. That convenience is real.

But convenience is not the same as protection.

A simple way to say it is this:

Cloud sync helps your files appear in more places.

A backup helps your files survive when something goes wrong.

Those two things overlap in the mind of the user. They do not always overlap in real life.

That is the problem.

This article is for people who use non-Apple computers, Windows laptops, office desktops, hospital computers, university computers, or cloud systems such as OneDrive, Google Drive, Dropbox, and Box. It is also for people who think they are “backed up” because their files are “in the cloud.”

That may be partly true.

But it may not be safe enough.

If your professional life depends on documents, the question is not whether your files appear in the cloud. The question is whether you can recover them when the cloud, the computer, or your own assumptions fail.

The rest of this article is the practical instruction and checklist I wish I had used before losing a month of work.

The yearly subscription pays off many times more of what you could lose without these instructions.

It may save you from learning the same lesson after the crash.

Read more

Subscribe now

In 30 years of clinical practice, I will encounter perhaps 6 of the 47 causes of lymphadenopathy I memorized in medical school. The other 41 occupy space where something more useful could live.

That is not a flaw in my memory. It is a design flaw in how we trained a profession.

For more than a century, medical education has rested on a single assumption: the doctor is the database. Every diagnosis, every drug interaction, every rare syndrome stored in one human brain. Exams rewarded recall. Residency rewarded the speed of recall.

The best doctor was the one who remembered the most.

For most of that century, this made sense. If you did not know it, you could not look it up at the bedside. The encyclopedia model was the only model available.

That era is over. And medicine has been slow to notice.

The Scale of the Problem

There are now more than 40,000 active biomedical journals. PubMed adds more than one million new citations every year. No physician has ever been able to keep up with this volume. We have known this for decades. We just pretended otherwise, and patients paid for the pretense with care shaped by what their doctor happened to have memorized last.

Share

Daniel Kahneman spent a career documenting how human judgment fails under conditions exactly like these. Physicians over-rely on what they have seen most recently. They anchor on the first diagnosis they consider. They remember dramatic cases and forget base rates. This is not a moral failure. It is how human cognition works under cognitive load and information overload.

AI does not have these biases. Used properly, it is not a replacement for physician judgment. It is a check on the ways physician judgment goes wrong.

What AI Actually Changes

Tools like Claude and ChatGPT can retrieve drug interactions, diagnostic criteria, guideline updates, and rare disease presentations in seconds, with citations. This does not replace the physician. It replaces the part of the job that was never really medicine: storage and retrieval of facts.

What medicine actually requires at its best is something different. A physician listens to a patient’s story. She notices what does not fit. She weighs risks against benefits for this specific patient, not the average patient in a trial population. She explains a difficult diagnosis to a frightened family. She makes a judgment call when the evidence is incomplete.

None of that requires memorizing 47 causes of lymphadenopathy. All of it requires presence, reasoning, and relationship. Those are the capacities medical training has historically crowded out in favor of encyclopedic recall.

When a physician does not have to spend half the appointment recalling whether a medication is safe in pregnancy, she can ask how the patient is actually doing. The doctors who embrace AI will not be lazier. They will be more present.

The Predictable Objection

Every time I raise this argument, someone says: doctors need to know things. You cannot just look everything up.

My father, who had to leave school at age 14, told me when I entered medical school: you do not need to remember everything, just know where to find the answer. That was in 1968, more than 50 years before ChatGPT.

He was right then. He is more right now.

Of course, core knowledge matters. A surgeon needs anatomy. An obstetrician needs fetal physiology. I am not arguing for ignorance. I am arguing for a different kind of knowledge: frameworks for thinking, not warehouses of facts. The ability to reason through a problem is not the same as the ability to recall a list.

What This Means for You

If your doctor uses AI-assisted tools to check a drug interaction or look up a rare presentation during your visit, that is not a sign of incompetence. It is a sign of good judgment. It means she is not relying on a memory formed years ago. It means she is checking current evidence for your specific situation.

The physicians who resist these tools are not protecting their patients. They are protecting the identity they built around memorizing things. These are not the same thing. Ask your doctor whether they use AI tools in your care. The answer will tell you something important.

Conclusion

I have practiced medicine for more than 50 years. I have forgotten more facts than most young physicians have yet learned. I am a better clinician now than I was at 40, not because I remember more, but because I know what questions to ask and how to reason toward an answer. AI handles the memory. The physician handles the medicine. These are not competing claims. They are a division of labor that medicine has needed for decades and that education has refused to acknowledge. The profession will come around. The technology does not wait for permission.

Bottom Line

Medical training built doctors into hard drives when it should have built them into processors. AI now gives us a better hard drive. The question is whether medical education has the courage to retrain for the job that has always actually mattered.

Medicine has always evolved through tools.

The stethoscope changed how physicians listened. Imaging changed how they saw. Electronic records changed how they documented and communicated. Each shift initially felt optional, even controversial. Over time, each became standard.

We are now in another transition. But this one is different.

Artificial intelligen…

Read more

Most physicians use AI like Google. There is a better way — and it takes one afternoon. I built it. Here is how it works — and what paid subscribers get to skip straight to.

Let me start with the thing nobody says: you cannot break anything. There is no wrong folder name, no wrong file, no configuration that locks you out or crashes something. The worst that happens is the system does not know who you are yet. That is fixable in five minutes.

I say this because I trained before computers existed, and I built this setup myself. If I can do it, so can you.

The setup has six steps. The first three are free. The last three — including the exact files I use and the one prompt that runs my entire workflow — are for paid subscribers.

Here is where we start.

First: Get the right app

Go to claude.com/download. Download the desktop app. Install it like any other program.

When it opens, look at the top of the screen. You will see three tabs: Chat, Cowork, and Code.

Click Cowork. That is where we are working.

If you have been using Claude in your browser, this is different. The browser version does not have Cowork. You need the desktop app.

Second: Make one folder with three folders inside it

On your computer — Desktop, Documents, wherever you keep things — create a new folder. Name it anything. I called mine “Claude Cowork.”

Inside that folder, make three more folders. Name them exactly: ABOUT ME, OUTPUTS, TEMPLATES.

That is the whole structure. You are done with this part.

Third: Understand what the three files do

Inside the ABOUT ME folder, you will put three short documents about yourself. The system reads them before every task — before you type a single word. That is what makes the output sound like you instead of like a press release written by someone who has never met you.

File one tells it who you are.

File two tells it what you never want to see.

File three tells it what you are working on right now.

Writing these files is the actual work of the setup. It takes about thirty minutes the first time. After that, you update them once a quarter.

What goes in them exactly — the structure, the language, the things that actually change the output — is what paid subscribers get below.

Below is everything you need to finish the setup in one sitting: the exact structure of all three files, word-for-word instructions for each one, the Global Instructions text ready to paste, and the first prompt I speak every morning to start my workflow.

If you have been thinking about upgrading, this is a good moment. The files alone will save you two hours of trial and error.

── For paid subscribers ──

Read more

Subscribe now

I write a medical study summary for ObGyn Intelligence roughly twice a week. The format is always the same: title line, 150-word summary, 150-word “What it means,” 150-word “My Take.” Seventh-grade readability. No jargon. Output as a Word document.

I used to type these instructions every time. Then I realized I was doing what I tell my residents never to do—repeating a task manually when a system could handle it. So I built a Skill. Now I say “Summ,” attach a PDF, and Claude produces the finished summary without a single additional instruction.

Every time.

Exactly right.

Skills, Cowork, and Scheduled Tasks are the features that let Claude work without you standing over it.

Here are 10 steps to set them up.

The 10 Steps

The rest of this post is for paid subscribers

Quick reminder on how to access Claude Cowork:

1. Go to claude.com/download. Download the app on your computer.

2. You must have a Pro account ($20/month). I pay for the $100/month plan.

3. Open the app. Click on the Cowork tab at the top between Chat & Code.

4. Select a folder from your computer. More about it right after this set up.

5. Make sure to always select “Opus 4.6” for complex tasks. It’s the smart model.

Step 1: Understand what a Skill is. A Skill is a set of detailed instructions that lives inside Claude and fires automatically when triggered. It is like a standing order in medicine. You do not re-write the insulin sliding scale every time—the standing order exists, the nurse follows it. A Skill works the same way. You define the task once, including the trigger, the format, the output, and the rules. Claude follows the standing order every time it is triggered.

Step 2: Identify your first Skill candidate. Think about the task you do most often with Claude. The one where you give the same instructions repeatedly. Summarizing articles. Drafting patient letters. Reviewing a study design. Writing social media posts about your practice. Whatever it is, if you have typed the same instructions more than twice, that task should be a Skill.

Step 3: Use Skill Creator to build it. You do not need to write code. Claude has a built-in Skill Creator that interviews you.

Open Cowork and type “Create a new skill.”

Claude will ask you questions:

What should this skill do? When should it trigger? What format should the output use? What rules should it follow? Answer the questions in plain language. Claude builds the Skill from your answers. It writes the instruction file for you.

Share

Step 4: Define your trigger word. Every Skill needs a trigger—a word or phrase that tells Claude to activate it. Make it short and specific. Mine is “Summ” for study summaries. Another is “NotesMeme” for creating quote graphics. Choose something you will not accidentally type in normal conversation. When Claude sees the trigger, it stops asking questions and starts executing.

Step 5: Write the negative triggers. This is the part that separates a good Skill from an annoying one. Negative triggers tell Claude when NOT to use the Skill. Without them, Claude may fire the Skill when you did not want it. If your Skill summarizes studies, tell it: “Do NOT use this skill when the user is asking a general question about a medical topic. Do NOT use when no PDF is attached.” Negative triggers prevent the Skill from hijacking conversations where it does not belong.

Step 6: Test the Skill with a real example. Attach a PDF or provide the input the Skill expects. Use the trigger word. Check the output. Is it the right length? The right format? The right tone? If not, go back and refine the Skill instructions. The first version is never perfect. The third version usually is.

Step 7: Open Cowork for multi-step tasks. Skills handle single, repeatable tasks. Cowork handles complex workflows. Cowork is Claude with access to a folder on your computer. It can read your files, create new documents, run calculations, and execute multi-step tasks without you guiding each step. If Skills are standing orders, Cowork is an attending running the service. You give the objective. It figures out the steps.

Step 8: Give Cowork a folder to work in. When you open Cowork, it asks you to select a folder on your computer. This is its workspace. Put the files you want it to work with in that folder—data sets, drafts, protocols, templates. Cowork reads them, processes them, and saves its output back to the same folder. You do not need to copy and paste anything in or out of the chat window. The work happens directly on your files.

Step 9: Set up a Scheduled Task. If you have a task that should run at the same time every day or week, Claude can schedule it. Open Cowork and say “Create a scheduled task.” Tell it what to do and when to do it. Example: every Monday morning, scan a specific folder for new PDFs and produce a summary document. As long as your computer is running, the task executes on schedule. You walk in and the work is already done.

Step 10: Build your second and third Skills. After your first Skill works, build another. And another. Each one removes a repetitive task from your workflow. I have Skills for study summaries, peer reviews, CDC data analysis, LinkedIn posts, and visual abstracts. Each one took about 15 minutes to build. Each one saves me 30 minutes or more every time it runs. After a month, the compound effect is substantial. You are not working harder. You are working once, and the system runs.

My Take

The difference between using Claude as a chat window and using it with Skills is the difference between dictating every order verbally and having a functioning EMR with order sets. Both get the job done. One of them scales.

I produce roughly eight times more content now than when I was still seeing patients. That is not because I work eight times harder. It is because the repetitive parts—formatting, structuring, finding the right template—are handled by Skills I built once and have not touched since. My time goes to the part that requires a physician: reading the study, forming the opinion, making the clinical judgment.

If you completed the steps from the previous post—the Project, the about-me file, the instructions—you already have a Claude that knows you. Now make it work for you. Build one Skill this week. You will build five more by the end of the month.

Every morning, the same ritual. You open Claude, type your question, and spend the first three messages explaining who you are, what you do, and what you need. You are an ObGyn. You want evidence-based answers. You do not want a disclaimer telling you to consult your physician. You are the physician.

Then you close the tab. Tomorrow, you do it again. Claude has no memory of you. It is like seeing a new locum every shift who has never read a single chart.

It does not have to work this way. Claude has features that remember you, your files, and your preferences across every conversation. Here are 10 steps to set them up. Total time: about 30 minutes. You will never re-explain yourself again.

The 10 Steps

Step 1: Open a Project. In Claude, click “Projects” in the left sidebar and create a new one. Give it a name that describes its purpose. Mine is called “CDC Natality Analysis.” Another is “ObGyn Intelligence Writing.” A project is a workspace. Everything inside it—files, instructions, conversations—stays together and carries forward. Think of it as a dedicated chart for a specific type of work.

Step 2: Write your Project instructions. Every project has a text box at the top for instructions. This is where you tell Claude how to behave inside this project. Be specific. Mine says things like: “Always use Vancouver citation format. Write at 7th-grade readability. Never fabricate references. When I say Summ, produce a 400-word summary in three sections.” These instructions apply to every conversation you start inside this project. Write them once. They persist.

Step 3: Upload your key files. Drag your most-used documents into the project. Practice protocols. Your favorite review articles. Patient education handouts. Clinical guidelines you reference often. Claude reads these files before answering your questions. If your gestational diabetes protocol is in the project, you can ask “What does my protocol say about insulin initiation?” and Claude will answer from your document, not from a generic training set.

Step 4: Create your about-me file. Open any text editor. Create a file called about-me.md. Write a short description of yourself: your specialty, your clinical focus, your research interests, your communication preferences. Mine says I am an MFM specialist and ethics consultant, that I use CDC natality data, that I write for a general audience at obmd.com, and that I do not tolerate hedging language. Upload this file to your project. Claude now knows who it is working with.

Subscribe now

Step 5: Be specific about what you do not want. This is the part most people skip, and it matters more than what you do want. Tell Claude what to avoid. “Do not add disclaimers about consulting a healthcare provider. Do not use the phrase ‘more research is needed.’ Do not soften conclusions when the evidence is clear.” Negative instructions are more powerful than positive ones. They prevent the generic AI voice from creeping back in.

Step 6: Start a conversation inside the project. Open your project and start a new chat. Ask a question you have asked before in basic chat. Notice the difference. Claude already knows your specialty, your preferences, and your files. It skips the preamble. It writes the way you told it to write. The answer is not generic—it is shaped by your context. This is the moment most physicians realize what they have been missing.

Step 7: Test it with a real task. Upload a study you recently read. Ask Claude to summarize it for a clinical audience. Or ask it to draft a patient counseling script based on your uploaded protocol. Or ask it to compare two approaches using the guidelines you provided. The more you test it with real work, the faster you learn what instructions to refine.

Share

Step 8: Refine your instructions after the first week. After five or six conversations, you will notice patterns. Claude does something you did not want. Or it misses a preference. Go back to your project instructions and edit them. This is not a one-time setup—it is a living document. I update my instructions roughly once a week. The AI gets sharper each time because you get more specific each time.

Step 9: Create a second project for a different purpose. Once you see the value, create another project for a different part of your work. One for research. One for patient education. One for administrative tasks. Each project has its own instructions, its own files, its own memory. They do not bleed into each other. Your research project can be precise and technical. Your patient education project can be warm and simplified. Same AI, different context.

Step 10: Set Global Instructions for things that apply everywhere. Claude also has a setting called Global Instructions—sometimes called “Custom Instructions” in Settings. These apply across all conversations and all projects. Use this for universal preferences: your name, your credentials, your department affiliation, your default citation format. Project instructions handle the specific. Global instructions handle the constant. Together, they mean Claude never starts from zero again.

My Take

I spent 50 years building clinical knowledge. The idea that I should re-explain my expertise to an AI every single session is absurd. These features exist to solve that problem. They are not advanced. They are not technical. They are a text box and a file upload. If you can write admission orders, you can configure a Claude project.

The physicians who tell me AI is not useful are almost always the ones using basic chat with no context, no files, and no instructions. That is not AI failing. That is a misconfigured tool. Fix the configuration. The results follow.

Part 2 of the series “You Are Using 5% of Your AI.” Next: The Features That Work Without You.

Subscribe to ObGyn Intelligence at obmd.com

Last Tuesday, I asked Claude to check my Google Calendar for open slots, draft an email to a colleague about a manuscript revision, and save the draft to Gmail. Claude did all three without me switching a single application. I stayed in one window. Claude moved between the tools.

That used to require opening three apps, copying information between them, and remembering to follow up. Now it requires one sentence.

This is what happens when Claude connects to the tools you already use. It stops being a separate application you visit and becomes the layer that runs across everything. Here are 10 steps to get there.

The 10 Steps

Step 1: Understand what Connectors are. Connectors are bridges between Claude and the applications you already use—Gmail, Google Calendar, Slack, Google Drive, and others. When a Connector is active, Claude can read from and act inside that application. It is not a separate login or a new platform. It is Claude reaching into your existing tools. Think of it as giving your research fellow access to your email and calendar instead of making her walk to your office every time she needs information.

Step 2: Connect Google Calendar first. Calendar is the easiest Connector to start with because the risk is low and the value is immediate. In Cowork, Claude will prompt you to connect your calendar when you ask about scheduling. Follow the authorization steps. Once connected, you can say “What is on my schedule Thursday afternoon?” or “Find me a free 30-minute slot this week” and Claude answers from your actual calendar—not a guess.

Step 3: Connect Gmail. Once Calendar works, connect Gmail. Now Claude can search your inbox, read specific emails, and draft replies. This is not Claude sending emails without your permission. It drafts them. You review. You send. But the drafting—finding the right thread, pulling relevant context, writing in your voice—that part is handled. For someone who writes 30 emails a day, this is not a novelty. It is a time savings you can measure.

Subscribe now

Step 4: Connect Google Drive. If your documents live in Google Drive, connecting it lets Claude search and read files stored there. You can say “Find the manuscript draft I worked on last week” and Claude locates it. Combined with your Skills, this means Claude can find a file, read it, and process it—all without you downloading anything or pasting text into a chat window.

Step 5: Set up Global Instructions. Global Instructions are preferences that apply across every conversation and every project. They are the constants in your workflow. Your name. Your degree. Your department. Your citation format. Your writing style. Set these once in Claude’s settings. They run forever, in every context. When Claude drafts an email, it signs with your correct credentials. When it writes a summary, it uses your preferred format. No reminders needed.

Step 6: Try Dispatch from your phone. Dispatch is Claude’s mobile feature. It lets you send instructions to Claude on your computer from your phone. You are at dinner and remember you need a summary of a paper sitting in your Cowork folder. Open Dispatch, type the instruction, and Claude executes it on your desktop. When you get home, the work is done. It is remote access to your AI workflow—not remote access to your computer, but to Claude running on it.

Step 7: Use Claude Computer for native desktop apps. Claude Computer is the feature that lets Claude see and interact with your Mac screen. It can click buttons, read what is displayed, scroll through documents, and navigate applications that do not have Connectors. If you need Claude to pull data from a desktop application that has no API—a hospital reporting dashboard, for instance, or a reference manager—Claude Computer is how it gets access. You grant permission for each application. Claude never acts on your screen without your approval.

Step 8: Combine Connectors with Skills. This is where the system becomes more than the sum of its parts. A Skill defines what Claude does. A Connector defines where it does it. Combine them: a Skill that summarizes a study and a Gmail Connector that drafts the summary to your co-author. A Skill that generates a patient handout and a Drive Connector that saves it to your shared clinical folder. A Skill that reviews your weekly schedule and a Calendar Connector that identifies conflicts. Each combination removes a manual step from your day.

Step 9: Build a Cowork Project for persistent memory. A Cowork Project is a workspace that persists between sessions. Your files stay. Your instructions stay. Your conversation history stays. Unlike regular Projects that reset context, a Cowork Project remembers what you worked on last time and picks up where you left off. Use this for ongoing work—a manuscript in progress, a research analysis that runs over weeks, a content calendar you update monthly. You do not restart. You continue.

Step 10: Audit your workflow and automate one more task each week. Now step back and look at your week. Where do you switch between applications? Where do you copy information from one place to another? Where do you do the same task on a schedule? Each of these is a candidate for a Connector, a Skill, or a Scheduled Task. You do not need to automate everything at once. One new automation per week. In two months, your workflow looks fundamentally different—not because you changed how you practice medicine, but because you stopped doing manually what a machine should handle.

My Take

I hear physicians say they do not have time to learn AI. This is exactly backward. You do not have time not to learn it. Every hour you spend configuring these tools saves you multiple hours every week for the rest of your career. The return on investment is not linear. It compounds.

The physicians who will thrive in the next decade are not the ones who know the most medicine. Medical knowledge is no longer scarce—Claude has access to more literature than any human can read in a lifetime. The physicians who will thrive are the ones who know how to use that knowledge efficiently. Who can direct an AI the way a senior attending directs a team: clearly, specifically, with the judgment that only experience provides.

You started this series using 5% of Claude. If you have followed all three posts, you are now using most of it. The chat window was the lobby. You have been given the keys to the building. Use them.

Part 4 of the series “You Are Using 5% of Your AI.” This is the final installment. Read the full series at obmd.com.

Subscribe to ObGyn Intelligence at obmd.com

A colleague asked me last month why she should pay for Claude when the free version works fine.

I asked her if she would practice obstetrics with only a stethoscope because the free version of a Doppler is her ears.

She laughed.

But she was serious, and so was I.

The free version of Claude is a demonstration. It lets you type questions and get answers. It uses a smaller, less capable model. It limits how many messages you can send per day. It does not give you access to Projects, Skills, Cowork, Connectors, or any of the features that turn Claude from a chatbot into a working system. Using free Claude and concluding that AI is not useful is like test-driving a car in a parking lot and deciding cars are slow.

This post walks you through getting the paid version, setting it up correctly from the start, and understanding why the investment makes sense. Ten steps. Twenty minutes. The beginning of a fundamentally different workflow.

The 10 Steps

Step 1: Understand what you are buying. Claude has three tiers. The free tier gives you limited access to a smaller model. Claude Pro costs $20 per month and gives you access to the most powerful model (Opus 4.6), extended thinking, Projects, file uploads, and significantly more usage. Claude Team costs $30 per user per month and adds collaboration features for groups. For a solo physician, Pro is what you want. Twenty dollars. That is less than one month of UpToDate, less than a single CME credit, and less than what most of us spend on coffee in a week. The comparison is not even close.

Step 2: Go to claude.ai and sign up. Open your browser. Go to claude.ai. Create an account with your email address. You will start on the free tier. That is fine. Use it for five minutes to see the basic interface. Then upgrade. Click on your profile icon, select “Upgrade to Pro,” and enter your payment information. The upgrade is instant. You immediately gain access to everything the free tier hides behind the wall.

Subscribe now

Step 3: Select the right model immediately. Once you are on Pro, the first thing you do is select the model. At the top of any new conversation, you will see a dropdown that lets you choose which version of Claude to use. Select Opus 4.6. It is the most capable model available. The default model (Sonnet) is faster but less thorough—fine for simple questions, but you are not paying $20 a month for simple questions. You are paying for the best reasoning available. Use it. Also turn on Extended Thinking in the same area. This tells Claude to reason step by step before answering. For anything clinical, analytical, or complex, the difference is substantial.

Step 4: Download the desktop app. Claude works in the browser, but the desktop app is better. Go to claude.ai/download and install it on your Mac or PC. The desktop app gives you access to Cowork—the feature that lets Claude read and create files directly on your computer. The browser version cannot do this. If you plan to use Claude for anything beyond typing questions into a chat window—and you should—the desktop app is where you want to be. It takes two minutes to install.

Step 5: Set your Global Instructions. Before you do anything else, go to Settings and find Global Instructions (sometimes called Custom Instructions). This is where you tell Claude who you are, once, for every future conversation. Write your name, your specialty, your credentials, your department. Add your preferences: “I am a board-certified ObGyn. Give me evidence-based answers without disclaimers about consulting a physician. Use Vancouver citation format. Be direct.” These instructions apply everywhere, in every conversation, from now on. You will never type them again.

Step 6: Create your first Project. Click “Projects” in the sidebar. Create a new project and name it after your most common use case. If you read studies regularly, call it “Literature Review.” If you write patient letters, call it “Patient Communication.” In the project instructions box, write specific directions for how Claude should behave in this workspace. Upload two or three files Claude should know about—a protocol, a guideline, a template you use often. This is the foundation. Every conversation inside this project will have this context automatically.

Step 7: Write your about-me file. Open any text editor on your computer. Create a file called about-me.md. Write a paragraph about yourself: what you do, what you care about, how you like information presented. It does not need to be long. Mine is about 200 words. It covers my specialty, my research focus, my writing platforms, and my communication preferences. Save it and upload it to your project. This single file transforms Claude from a generic AI into one that responds in a way shaped by your expertise and priorities. Every physician I have shown this to says the same thing: “Why did I not do this months ago?”

Step 8: Test it with real work. Do not test Claude with toy questions. Open your project and give it a real task. Upload a study you need to read. Ask it to summarize a guideline. Have it draft a letter you actually need to send. Compare the output to what you got from free Claude with no context. The difference is not subtle. Free Claude gives you a generic answer that could be for any physician. Configured Pro Claude gives you an answer that is for you, using your files, following your rules, in your preferred format. This is the moment the $20 makes sense.

Step 9: Understand what you are not getting with the free version. This is worth stating plainly. The free tier does not give you Opus 4.6—you are stuck with the less capable model. It limits your messages to roughly 10–20 per day before it throttles you or downgrades the model further. It does not support Projects with file uploads. It does not give you Cowork, Skills, Connectors, Scheduled Tasks, or Claude Computer. It does not support Extended Thinking. In practical terms, the free tier is a text box that forgets you. The paid tier is a configurable system that learns your preferences, reads your files, and works across your tools. These are not the same product. They share a name.

Step 10: Commit to one week. Use Claude Pro for one full week. Use it every day, for real tasks, inside your project, with your about-me file loaded. At the end of the week, add up the time it saved you. For most physicians, the answer is somewhere between two and five hours. That is $20 for two to five hours of your time back. Every week. For comparison, the average ObGyn’s hourly rate is well north of $100. If Claude saves you even one hour per week, the subscription pays for itself five times over. After the first week, you will not debate whether it is worth it. You will wonder why you waited.

My Take

Physicians pay for UpToDate without blinking. They pay for society memberships, MOC fees, CME courses, and journal subscriptions. They accept that professional tools cost money. Then they try the free version of an AI that is more powerful than all of those tools combined, get a mediocre result, and say AI is overhyped.

It is not overhyped. You are undertooled.

The free version of Claude is a brochure. The Pro version is the product. You would not evaluate a new ultrasound machine by looking at the brochure. You would turn it on, scan a patient, and judge the image quality. That is what Step 8 is for. Try the real thing with real work. Then decide.

Twenty dollars a month. The best investment in your practice you will make this year.

A doctor colleague told me last week she uses Claude every day. I asked her how. “I type a question and it gives me an answer.”

That is it.

That is how she uses it.

She is a board-certified ObGyn running a busy practice.

An MFM.

And she is using the most powerful AI tool available the same way she would use Google in 2005.

She is not alone. Roughly 95% of physicians who use AI stop at the chat window. They ask a question, get a response, close the tab, and next time start from zero. The AI forgets them. They forget the AI can do more. Both sides lose.

This post is a plain-language guide to what Claude can actually do, written for ObGyns who have used the basic chat but nothing else. No coding. No technical background. Just the features that matter, explained in the order you should learn them.

What You Already Know

Chat is the basic window. You type, Claude answers. Think of it as texting a very smart stranger who has never met you before and will forget you the moment you close the conversation. Every session starts fresh. No memory, no context, no relationship.

Extended Thinking is a setting that tells Claude to reason more carefully before answering. Instead of giving you the first plausible response, it works through the problem step by step. For anything clinical or analytical, turn it on. It takes a few extra seconds. The answers are meaningfully better.

Opus 4.6 is Claude’s most capable model. Always select it. The default model is faster but less thorough. You would not send your resident to handle a complicated delivery when your MFM specialist is available. Same principle.

Subscribe now

The Features That Remember You

This is where most physicians never go, and it is where the real value starts.

• Projects are workspaces where you upload files and write instructions that Claude remembers across every conversation in that project. I have a project for CDC natality data analysis. Every time I open it, Claude already knows my datasets, my variable definitions, and my analytical preferences. I do not re-explain anything. Think of Projects as hiring the same research assistant every day instead of training a new one each morning.

• Context Files are documents you drop into a project so Claude can draw on them. Your practice protocols. Your patient education handouts. Your favorite review articles. The more context you give, the less prompting you need. Claude reads them before it reads your question.

• About-me.md is a single file that tells Claude who you are, what you do, and how you think. Your specialty. Your clinical interests. Your preferred communication style. Your pet peeves. Write this once, and Claude stops giving you generic answers. It gives you answers shaped by your expertise and priorities. This one file changes everything.

  Share

The Features That Work Without You

• Skills are sets of detailed instructions that live inside Claude and fire automatically when triggered. I have skills for summarizing medical studies, writing peer reviews, analyzing CDC data, and drafting LinkedIn posts. When I say “Summ” and attach a PDF, Claude does not ask me what I want. It already knows: 400–500 words, three sections, 7th-grade readability, output as a Word document. Every time. Without being told twice.

• Skill Creator is Claude’s built-in tool that interviews you about a process and then generates the skill for you. You do not need to write code. You describe what you want, answer a few questions, and Claude builds the instructions. If you find yourself giving Claude the same directions more than twice, you need a skill.

• Cowork takes this further. It reads your files, creates documents, runs code, and executes multi-step tasks. Think of it as the difference between dictating a letter and handing someone your entire filing cabinet and saying “prepare the quarterly report.” Cowork can access a folder on your computer, work through the files, and produce deliverables. I use it daily.

• Scheduled Tasks let Claude run jobs automatically on a schedule. Set it once, and it runs without you. As long as your computer is on, the task executes at the time you specified.

The Features That Connect Everything

Connectors plug Claude directly into your existing tools. Slack, Google Calendar, Gmail. Claude does not just answer questions about your schedule—it reads your calendar, drafts emails, and acts inside the applications you already use.

Global Instructions are persistent instructions that apply across every conversation, not just within a single project. Set them once. They run forever. Mine tell Claude my department affiliation, my writing standards, and where to save files. I never repeat these instructions.

My Take

I have been writing about AI in medicine since before most professional societies acknowledged it existed. The technology is not the bottleneck. The bottleneck is physicians who open the basic chat, get a mediocre answer to a poorly framed question, and conclude that AI is overhyped.

It is not overhyped. It is underconfigured.

The difference between basic chat and a fully configured Claude is the difference between asking a random medical student a question in the hallway and working with a trained research fellow who knows your data, your preferences, and your standards. The fellow does not just answer—she anticipates. That is what a configured AI does.

You do not need to learn all of this at once. Start with one thing: create a Project, upload your most-used protocols, and write a short about-me file. Use it for a week. You will not go back to the basic chat.

The 95% of Claude you have not touched is where the actual value lives.