site banner
PaperclipPerfector  1d ago (text post)   1293 thread views

Small-Scale Question Sunday for June 7, 2026

Do you have a dumb question that you're kind of embarrassed to ask in the main thread? Is there something you're just not sure about?

This is your opportunity to ask questions. No question too simple or too silly.

Culture war topics are accepted, and proposals for a better intro post are appreciated.

2
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

What employees have you fired so far?

In Zvi's recent post, I noticed an interesting pairing of two things:

Sell your house. Stuart Thompson lets Gemini (because he had a free account there from work that saved him $8 a month?!) walk him through everything involved in the sale, including being his agent. The problem is, Stuart does not seem to realize he does not know the counterfactual?

Stuart A. Thompson: In the end, using A.I. netted me more than $90,000. That includes the premium over the asking price, plus the roughly $36,000 in fees I didn’t pay.

I mean, yes, the agents he talked to early on told him he’d lose money, and instead he turned a profit. But only after the sale did he talk to another agent for an expert opinion, and that expert expected a higher sale price than Stuart got, meaning he almost certainly listed too low. Stuart thinks that after the agent fee he still basically broke even, but I’m guessing he put in more work and stress this way, and took on more downside risk. I know that if I am ever selling or buying, I will be using AI extensively as part of the effort, but I am going to stick with Danielle Wiedemann. I am confident that her help, connections and advice were worth far more than the fee, and would be again.

and:

For those confused about the radiology example, yes, AI is better than radiologists at reading x-rays, and many other components of professional services, and does so at cost epsilon, and this is super useful. Even if no one is out of work quite yet, often there is a ton of value in ‘pretty good answer, vastly better than you could otherwise get without a professional, for cost ~$0’ when the professional costs $1,000 and up.

It was a bit stark, because getting a pretty good answer, vastly better than you could otherwise get without a real estate professional, would seem to cost ~$0, when the professional apparently costs something like $36,000 and up. So why not fire the real estate agent?

There could be a variety of reasons involving the nature of the work, regulatory barriers, etc., but one thing that comes to mind is that Zvi has paid for a real estate agent before and is consciously thinking about what that situation is like when thinking about whether he would hire again. Whereas, I doubt he's hired a radiologist before and is probably not in a situation where he's thinking super seriously about the considerations that would be involved if he had a need for such a service.

This leads me to ask, "Which employees have you fired?" In this case, "employees" can be read broadly, covering folks like real estate agents/radiologists, who you may procure services from on occasion, in addition to actual employment relations if you're a manager/business owner. But I want to particularly hone in on examples where you have paid a human for a particular service in the past and have subsequently encountered a nearly-identical need, but have chosen to not pay a human now for the service.

This question is in significant part simply selfish. I might be missing some aspect in my life where I can save a bunch of money. That would be cool, and I'd like to do that if I can.

re: Stuart Thompson's article, I was surprised he purchased his "ranch" for around $500K in upstate New York, a real estate market that's been hot since Covid, on a journalist's salary. He walked away with about $600-700K from the sale, according to the article, and I doubt he can afford a bigger house in the same area.

I think I'd go for the house selling rather than the radiologist replacement. Big difference between "dang, priced it too low, I could have made another $20,000" and "dang, AI missed the tumour that's going to kill me".

The fundamental problem with AI in medicine is that even if you can use AI for self diagnosis you still can't do anything about it without seeing a professional. It's a bit like still having to hire a realtor but now you get to suggest a target price.

You shouldn't use either an AI or a listing agent in this market. What "connections" do you need to sell a house?

I haven't fired anyone, but I've probably reduced their hours while increasing the service I get. I didn't fire my mechanic, but I'm often able to diagnose problems more effectively and go into my mechanic saying "x is wrong" rather than "idk man you tell me." I haven't fired my engineer, but when I have a casual question on regulations I ask chatgpt and it points me to the relevant code section, rather than paying my engineer to do so.

I use an LLM before every doctor's visit to get myself familiar with the issue so that I know what exactly to talk about to my doctor. Based on my experience, as long as it's not urgent care or a serious issue, an average (specialized) doctor's visit lasts about 7-10 minutes, where most of the time is you answering some questions for the nurse. Then the main doctor comes in for 2-3 minutes and that's it, here's your $200-400 bill. So, to get any real help, I need to come in with a clear understanding of what my issue is and what type of treatment I'm seeking, that way I can direct the conversation and get actual help. One time I was vindicated with my approach when I was about to be prescribed a drug with a common side effect of drying out meibomian glands (eyes), which would've definitely left me with permanent damage and lifelong pain had I not researched my issue, learned that this drug is a common solution to it and that this was it's main side effect beforehand. I had a laser eye correction surgery few months prior, and FDA has an advisory notice to not take this drug for at least 6 months before/after laser eye surgery. The doctor was dumbfounded when I told her about it, genuinely seemed like it was her first time hearing of this. A LLM didn't have access to my full medical history, so it didn't directly bring this up either, but I was able to connect the dots and research further once I learned that one of the side effects had to do with drying out your eyes. I think you could easily replace large chunk of all doctor visits with an LLM and get comparable, perhaps even better level of care. But I doubt it ever happens, healthcare is a big pie and beneficiaries will fight to keep it as is.

I used AI to replace what I guess could have been a doctor's visit. I had some questions about an OTC medication I was taking for seasonal allergy issues, so I went to Gemini. The initial question seemed a bit basic to bother setting up an actual doctor's visit. It gave good answers to everything, helped identify some issues it was causing me, and suggested some other treatments (other OTC medications), one of which I tried and was much better. It's honestly a way, way better experience than anything I've done with the "proper" healthcare system. AI will give you an answer to any sane question at any hour of the day or night immediately, doesn't mind at all if I ignore a conversation for a week and then ask another question to continue it, and it actually listens to things you tell it better than most doctors. I know it's not as reliable as a real doctor, but I think I'm smart enough to apply common sense, ask for references to properly validated information sources, and only actually use treatments purchased from a real brick-and-mortar drugstore that are advertised to treat the proper conditions and used in accordance with their labeling etc.

In comparison, a proper doctor requires the slog of finding one of the appropriate specialty that my insurance accepts, setting an appointment that might be a week or two away, travelling to the office, usually filling out a ton of forms answering a slew of questions unrelated to my actual complaint, usually getting seen late, talking to a nurse first, then actually talking to a doctor for a few minutes, and that's the only time they'll actually answer any questions, since apparently email is too insecure, so nobody uses it, even if I positively give zero fucks who knows about my allergy issues. Though even if the system worked perfectly and I could use email or some similarly convenient modern communication medium to ask questions of a doctor, it's a bit much to expect a real expert to answer any question instantly 24x7. I know there are some things that will basically always require a real doctor, but for the minor stuff in the gaps, this feels like a great solution to me.

My experience with ai medical is the same reason I wouldn't dare use it for therapy. It will indeed let you just keep talking and asking questions forever in a way your real doctor will not. Which is fine when I need answers about my hip hurting (probably a muscle strain from my move, thanks chatgpt, and it is getting better now that I'm trying to avoid certain activities, thanks chatgpt) but I could imagine obsessing about a thing that was emotionally upsetting me for an unhealthy long time. With a person you are concerned about their opinion of you and that trims the excesses of narcissism, as does the ticking away of an expensive hour.

The AI radiologist has the same problem as the self-driving car and the AI lawyer. Tech companies can tout their products all they want, but unless they're a value add for a radiologist who isn't getting fired, they aren't going to be adopted. Whenever you hire a professional to do something, part of the cost is paying for the work, but another part of the cost is paying for someone else to be responsible for the work. I can do a lot of things on my own car, but I'm not touching anything that can cause catastrophic damage if I fuck it up. Even if taking it to a mechanic doesn't reduce this risk, if the mechanic fucks up they're responsible for it. Same thing with the radiologist—if it misses a serious problem who is going to take responsibility in a malpractice suit? The hospital? The tech company? Considering that both have spent decades vehemently denying that they have any responsibility for how end-users use their products or licensed professionals under their employ, as the case may be, I doubt either of these is going to change their position just because of AI. The first lawsuit will likely suck up whatever cost savings they see, and I doubt the carriers would be on board.

Lawyers and doctors are a rather special case though (and self-driving cars appear to have been solved given Waymo's rapid expansion). When a company hires a product manager, a customer support person, an executive assistant, a marketing executive, a salesperson, a software developer, or the myriad other back office roles that fill a standard corp, they are not especially worried about potential liabilities (financial roles excepted). You don't need many of those roles to be replaced before serious economic consequences become apparent, and the protection of law or medicine is rendered moot.

That being said, I do think there will be a lot of inertia just by the nature of corporations. Most big corpos are already too big and could probably have slashed 20% of their workforce even before AI. It's part of the nature of the firm and not something easily shaken off.

radiologist

I recently got a cheap knee MRI at a local facility and wanted to double-check the radiologist's report and maybe learn something about how to read MRIs. So I uploaded the .zip of obscurely formatted images to Claude (highest-powered free model as of a week ago, on high setting) and asked for an opinion. It did a great job of parsing the file and coming up with a plausible-sounding stream of orthopedic gobbledegook, but failed to identify a partially torn quadriceps tendon, which was far and away the most significant finding in the radiologist's report and is generally consistent with symptoms and history. When I nudged it a little bit, it seemed to pick up on the quad tendon, but when I asked it to highlight the finding on the relevant frame it identified a random location on the IT band as the quad tendon. I wasn't quite sure myself at first, not having a ton of experience with either AI or reading MRIs, and I got Claude fairly confused by asking about the difference between sagittal and coronal before I figured out what was going on. (All this took several days, since I was hitting my token cap fairly quickly on high setting.). Super impressive natural language processing, but I'm not sold on its radiological abilities, and I seem to recall hearing quite a while ago that the days of human radiologists were numbered.

Yeah I am sure that AI will come for Radiologists and everyone else eventually but we are still far away. They aren't right all the time, and the expectation is you are right all the time for one (and when they are wrong they are sometimes wrong in the most basic of ways).

That says nothing about being about to talk to primary teams about the read, rare findings, and how to manage incidental findings, as well as other basic radiology tasks.

People want this fight to be over because they hate doctors or certain kinds of expense, but we aren't there yet.

I wonder how its radiological abilities compare to ultrasonic nondestructive testing analysis? Seems like a similar field, and I believe it was getting touted for the latter as well.

It looked like much of the skill of ultrasound specialists was in moving the device around to get the correct locations in real time, and then they would mark it off on their checklist, and a doctor would look at it later (but also they already have an opinion on morality in the moment). So an LLM could replace the analysis, but not the moving the device around while jostling the item being scanned (at least for pregnancies).

Yeah, I was thinking much more of the interpretation that the initial reading-taking.

I mean, yes, the agents he talked to early on told him he’d lose money, and instead he turned a profit. But only after the sale did he talk to another agent for an expert opinion, and that expert expected a higher sale price than Stuart got, meaning he almost certainly listed too low. Stuart thinks that after the agent fee he still basically broke even, but I’m guessing he put in more work and stress this way, and took on more downside risk. I know that if I am ever selling or buying, I will be using AI extensively as part of the effort, but I am going to stick with Danielle Wiedemann. I am confident that her help, connections and advice were worth far more than the fee, and would be again.

Quite bogus. The second agent was told that the house was sold at the price it did with the help of AI. He gets zero points for scoffing and saying he would have totally gotten a better price for it.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats