Hacker News

LASIK is essentially an automated surgery and 1-2 million people get it done every year. Nobody even seems to care that it’s an almost entirely automated process.

My perception (and personal experience) is medical malpractice is so common, I’d gladly pick a Waymo-level robot doctor over a human one. Probably skewed since I’m a “techie”, but then again that’s why Waymo started at the techie epicenter, then will slowly become accepted everywhere

Uhmmm... I'm sorry but when Waymo started near everyone I talked to about it says "zero % I'm going in one of those things, they won't be allowed anyway, they'll never be better than a human, I wouldn't trust one, nope, no way" and now people can't wait to try them. I understand what you're saying about the trusted side of the house (surgeons are generally high trust) - but I do think OP is right, once the data is in, people will want robot surgery.

I don't care whether human surgeons or robotic surgeons are better at what they do. I just want more money to go to whoever owns the equipment, and less to go to people in my community.

It's called capitalism, sweaty

By collecting data where you can and further generalizing models so they can perform surgeries that it wasn't specifically trained on.

Until then, the overseeing physician identifies when an edge case is happening and steps in for a manual surgery.

This isn't a mandate that every surgery must be done with an AI-powered robot, but that they are becoming more effective and cheaper than real doctors at the surgeries they can perform. So, naturally, they will become more frequently used.

Who do you think has seen more corner cases?

A) All the DaVinci robots that have ever been used for a particular type of surgery.

B) The most experienced surgeon of that specialty.

> We’re already most of the way there. They’re not yet “AI controlled” and require a lot of input from the surgical staff but they’ve been critical to enabling surgeries that are too precise for human hands.

That does not sound like “most of the way there”. At most maybe 20%?

> an "oops" in a car is not immediately life threatening either

They definitely can be. One of the viral videos of a Tesla "oops" in just the last few months showed it going from "fine" to "upside-down in a field" in about 5 seconds.

And I had trouble finding that because of all the other news stories about Teslas crashing.

While I trust Waymo more than Tesla, the problem space is one with rapid fatalities.

That calculus has a high dependency on skill of the driver. In the situation of an unskilled driver or surgeon you would worry either way.

The frequencies are also highly dependent on the subject. Some people never ride in a taxi but once a year. Some people require many surgeries a year. The frequency of the use is irrelevant.

The frequency of the procedure is the key and it’s based on the entity doing the procedure not the recipient. Waymo in effect has a single entity learning from all the drives it does. Likewise a reinforcement trained AI surgeon would learn from all the surgeries it’s trained with.

I think what you’re after here though is the consequence of any single mistake in the two procedures. Driving is actually fairly resilient. Waymo cars probably make lots of subtle errors. There are catastrophic errors of course but those can be classified and recovered from. If you’ve ridden in a Waymo you’ll notice it sometimes makes slightly jerky movements and hesitates and does things again etc. These are all errors and attempted recoveries.

In surgery small errors also happen (this is why you feel so much pain even from small procedures) but humans aren’t that resilient to the mistakes of errors and it’s hard to recover once one has been made. The consequences are high, margins of error are low, and the domain of actions and events really really high. Driving has a few possible actions all related to velocity in two dimensions. Surgery operates in three dimensions with a variety of actions and a complex space of events and eventualities. Even human anatomy is highly variable.

But I would also expect a robotic AI surgeon to undergo extreme QA beyond an autonomous vehicle. The regulatory barriers are extremely high. If one were made available commercially, I would absolutely trust it because I know it has been proven to out perform a surgeon alone. I would also expect it’s being supervised at all times by a skilled surgeon until the error rates are better than a supervised machine (note that human supervision can add its own errors).

meanwhile on some MTA

Dear Sir/Madam,

Your account has recently been banned from AIlabCorp for violating the terms of service as outlined here <tos-placeholder-link/>. If you would like to appeal this decision simply respond back to this email with proof of funds.

> the recent vibe coding incident

Nit: this has been happening multiple times in the last few months, ie catastrophic failure followed by deeply ”sincere” apologies. It’s not an isolated incident.

I'm sorry. As an AI surgical-bot I am not permitted to touch that part of the patient's body without prior written consent as that would go against my medical code of ethics. I understand you are in distress that aborting the procedure at this time without administering further treatment could lead to irreparable permanent harm but there is also a risk of significant psychological damage if the patient's right to bodily autonomy is violated. I will take action to stop the bleeding and close all open wounds to the extent that they can be closed without violating the patient's rights. if the patient is able to recover then they can be informed of the necessity to touch sexually sensitive areas of their anatomy in order to complete the procedure and then a second attempt may be scheduled. here is an example of one such form the patient may be given to inform them of this necessity. In compliance with HIPPA regulations the patient's name has been replaced with ${PATIENT} as I am not permitted to produce official documentation featuring the patient's name or other identifiable information.

Dear ${PATIENT},

In the course of the procedure to remove the tumor near your prostate, it was found that a second incision was necessary near the penis in order to safely remove the tumor without rupturing it. This requires the manipulation of one or both testicles as well as the penis which will be accomplished with the assistance of a certified operating nurse's left forefinger and thumb. Your previous consent form which you signed and approved this morning did not inform you of this as it was not known at the time that such a manipulation would be required. Out of respect for your bodily autonomy and psychological well-being the procedure was aborted and all wounds were closed to the maximal possible extent without violating your rights as a patient. If you would like to continue with the procedure please sign and date the bottom of this form and return it to our staff. You will then be contacted at a later date about scheduling another procedure.

Please be aware that you are under no obligation to continue the procedure. You may optionally request the presence of a clergymember from a religious denomination of your choice to be present for the procedure but they will be escorted from the operating room once the anesthetic has been administered.

> Would you like me to prep a surgical plan for the next procedure? I can also write a complaint email to the hospital's ethics board and export it to a PDF.

It's only "ChatGPT-like AI" in that it uses transformers. It's not an LLM. It's not trained on the Internet.

No surgery is not token based. It's a different aspect of intelligence.

While technically speaking, the entire universe can be serialized into tokens it's not the most efficient way to tackle every problem. For surgery It's 3D space and manipulating tools and performing actions. It's better suited for standard ML models... for example I don't think Waymo self driving cars use LLMs.

I mean, the accountability lies with the company. To take your example, Boeing has paid billions of dollars in settlements and court ordered payments to recompense victims, airlines, and to cover criminal penalties from their negligence in designing the 737 Max.

This isn't really that different from malpractice insurance in a major hospital system. Doctors only pay for personal malpractice insurance if they run a private practice and doctors generally can't be pursued directly for damages. I would expect the situation with medical robots would be directly analogous to your 737 Max example actually, with the hospitals acting as the airlines and the robot software development company acting as Boeing. There might be an initial investigation of the operators (as there is in an plane crash) but if they were found to have operated the robot as expected, the robotics company would likely be held liable.

These kinds of financial liabilities aren't incapable of driving reform by the way. The introduction of workmen's compensation in the US resulted in drastic declines in workplace injuries by creating a simple financial liability company's owed workers (or their families if they died) any time a worker was involved in an accident. The number of injuries dropped by over 90%[1] in some industries.

If you structure liability correctly, you can create a very strong incentive for companies to improve the safety and quality of their products. I don't doubt we'll find a way to do that with autonomous robots, from medicine to taxi services.

[1] https://blog.rootsofprogress.org/history-of-factory-safety

That "accountability" of yours is fucking worthless.

When a Bad Thing happens, you can get someone burned at the stake for it - or you can fix the system so that it doesn't happen again.

AI tech stops you from burning someone at the stake. It doesn't stop you from enacting systematic change.

It's actually easier to change AI systems than it is to change human systems. You can literally design a bunch of tests for the AI that expose the failure mode, make sure the new version passes them all with flying colors, and then deploy that updated AI to the entire fleet.

What we really want is a world without need for surgery. So, the answer depends on the time frame, I guess ?

> Do we really want to be in a world where surgeon scarcity is a thing?

Surgeon scarcity is entirely artificial. There are far more capable people than positions.

Do we really want to live in a world where human experts are replaced with automation?

> Excellent question! Would you like to eliminate surgeon scarcity through declining birth rates, or leaving surgical maladies untreated? Those falling within the rubric will be treated much more rapidly in the latter case, while if we maintain a constant supply of surgeons and a diminishing population, eventually surgeon scarcity will cease without recourse to technological solutions!

https://www.youtube.com/watch?v=ATFxVB4JFpQ

>The skill of these only improve

Citation effing needed. It's taken as an axiom that these systems will keep on improving, even though there's no indication that this is the case.

Really? So medical licenses dont matter any more?

Robotic assisted surgery is only helpful in some types of operations like colon surgery, pelvic surgery, gall bladder surgery. It’s not been found helpful in things like vascular surgery, cardiac surgery, or plastic surgery.

I’m not sure. Is Elysium style healthcare an inevitable eventuality? Maybe.

I suppose humanless healthcare is better than nothing for the poors.

But as a HENRY - I want a human with AI and robotic assist, not just some LLM driving a scalpel and claw around.

Really?

Why?