Hacker News Reader

Comment by polski-g | Hacker News Reader

[flagged]

The limits of the Model T were both: "well defined" and "deterministic". LLMs are neither.

It's interesting how "AI" has become synonymous with "LLM" in people's minds.

There are many classifier models that have real value add. Catching potential mistakes or omissions in analysis has been well received (by docs and admins).

Be a tool to help doctors do a better job, help them deal with their increasing workloads.

When 100% of liability for malpractice is placed upon the senior execs and major shareholders of the companies pushing medical AI my skepticism will abate slightly. Until that happens you've got perverse incentives and no downsides to provide meaningful guardrails.

Usually when something's value and limitations has been proven people stop referring to it as AI. AI is used as a marketing term to generate hype for new thing.

Do you actually believe that is an apt comparison?

> “No computer, no AI can replace a human touch,” said Amy Grewal, a registered nurse.

I don't believe this to be true in the long run. We will almost certainly have Westworld like robots in the future. Researchers have also found, on average, ChatGPT has better bedside manners than doctors.

Bedside manner, you say? Get back to us with it's clinical success rates.

If you stop being in the replace mindset, you could see the following happen...

1. Patient gets tests / imaging

2. Doctor does analysis

3. AI double checks

4. Doctor writes final report

5. AI communicates with more empathy

If you haven't had a dbag doctor, you might not know how an AI might be preferable to that experience

Step 3.5 Doctor rewrites analysis to go with whatever the AI spits out because:

* Too busy (assumptions were made that patient load could increase with ML support)

* Too scared of being sued even if the doctor disagrees

* Too scared of being fired because the doctor keeps overriding the "cost effective" ML verdicts

We've seen this play out over and over in the medical field. I've watched my Primary Care physician go from angst over patient outcomes to angst over statistical outcomes for his department. It took 15 years, but the system got him.

I'd rather have a dbag doctor than a computer with fake emotions.

Everyone has preferences, many people will prefer fake emotions over condescendence or elitism

People should have a choice

True and if they're leading with their fucking feels when making major life decisions their arrested emotional development is liable to land them in deeply regrettable circumstances.

... even if the tech materializes, is Westworld the thing you want to reference if you're arguing that AI replacing humans is a good thing? I think all incarnations of that book/film/show involve robots that (a) don't work as intended and (b) kill people (c) make humans behave badly.

I'm using the lifelikeness aspect of Westworld (or whichever scifi you prefer), more than the mental drives of the AI

Sex bots are around the corner and driving the progress.

I'd be happy with a Star Wars droid for a doctor if it meant it cost less and I didn't spend more time waiting than being helped

It took a century of blood to make cars as safe as they are today.

A quick glance at crash statistics for the US suggests that no meaningful advances have been made in automobile safety in the last 20 years give or take. This assumes you accept fatalities per driver hour as a valid metric.

You have to realise the cost function being optimised isn't "lives saved", but lawsuits Vs profits for shareholders.

I feel like we're arguing the same side here? We didn't get mandatory in-console ADHD factories and 6000lb daily drivers that punch through DOT guardrails because they enhance safety.

I honestly hope they make you drive the half-baked Model T.

AI doctor can right now run laps around your PCP and Urgent Care docs so I'm not sure what the point you're trying to make is.

AI doctor right now also doesn't suggest that all my medical issues are "are you sure you're not on your period?"

Reply on news.ycombinator.com