This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
Taking this at absolute face value, I wonder if this is at least partially because the specialists will have observed/experienced various 'special cases' that aren't captured by the medical literature and thus aren't necessarily available in the training data.
As I understand it, the best argument for going to an extremely experienced specialist is always the "ah yes, I treated a tough case of recurrent Craniofacial fibrous dysplasia in the Summer of '88, resisted almost all treatment methods until we tried injections of cow mucus and calcium. We can see if your condition is similar" factor. They've seen every edge case and know solutions to problems other doctors don't even know exist.
(I googled that medical term up there just to be clear)
LLMs are getting REALLY good at legal work, since EVERYTHING of importance in the legal world is written down, exhaustively, and usually publicly accessible, and it all builds directly on previous work. Thus, drawing connections between concepts and cases and application to fact patterns should be trivial for an LLM with access to a Westlaw subscription and ALL of the best legal writing in history in its training corpus.
It is hard to imagine a legal specialist with 50 years of experience being able to outperform an LLM that knows all the same caselaw and law review articles and has working knowledge of every single brief ever filed to the Supreme Court.
I would guess a doctor with 50 years of experience (and good enough recall to incorporate all that experience) can still make important insights in tough cases, that would elude an AI (for now).
I’m in a very specialized area of law. While there is a lot of law, you’d be surprised daily how many fact patterns I face where there is no guidance (either judicial, administrative, or secondary) and things fall down at the edges (ie basically comes down to judgement).
Moreover, law changes all of the time (especially in this field). This seems to confuse LLMs sometimes (both in what the current law is and what the change in law means and doesn’t mean). Finally, a lot of the guidance doesn’t strictly apply in one area but can (taking into account a lot of factors) apply to totally different area without any indication.
Further, my role isn’t primarily telling what the answer is but figuring out what the facts are, what they can be, and what the best set of future facts are applied to an unclear legal framework whilst trying to predict future government policy.
We’ve tried using LLMs. They’ve all failed to this point.
I mean, yeah, if a legislator passes a big, comprehensive new package that revamps entire statutes then there's no readily applicable case law, then its anybody's game to figure out how to interpret it all, an experienced attorney might bring extra gravitas to their argument... I'm not sure they're more likely to get it right (where 'right' means "best complies with all existing precedent and avoids added complexity or contradictions," not "what is the best outcome for the client.")
But this is my point. If you encounter an edge case that hasn't been seen, but have a fully fleshed-out fact pattern and access to the relevant caselaw (identifying which is relevant being the critical skill) why would we expect a specialist attorney to beat an LLM? Its drawing from precisely the same well, and forging new law isn't magic, its using one's best judgment, balancing out various practical concerns, and trying to create a stableish equilibrium... among other things.
What really makes the human's judgment more on point (or, the dreaded word, "reasonable") than a properly prompted LLM's?
I've had the distinct pleasure of drilling down to finicky sections of convoluted statutes and arguing about their application where little precedent exists. I've also had my arguments win on appeal, and enter the corpus of existing caselaw.
ChatGPT was still able to give me insightful 'novel' arguments to make on this topic when I was prepping to argue a MSJ on this particular issue by pointing out other statutory interactions that bolster the central point. It clearly 'reasons' about the wording, the legislative intent, the principles of interpretation in a way that isn't random.
Also, have you heard of the new law review article that argues "Hallucinated Cases are Good Law." It argues that even though the AI is creating cases that don't exist out of whole cloth, they do so by correlating legal concepts and principles from across a larger corpus of knowledge and thus they're hallucinating what a legal opinion "should" be if it accounted for all precedent and applied legal principles to a given fact pattern.
I find this... somewhat compelling. I don't think I've encountered situations where the AI hallucinated caselaw or statutes that contradicted the actual law... but it sure did like to give me citations that were very favorable to my arguments, and phrased in ways that sounded similar to existing law. Like it can imagine what the court would say if it were to agree with my arguments and rule based on existing precedent.
I dunno. I think I'm about at the point where I might accept the LLM's opinion on 'complex' cases more readily than I would a randomly chosen county judge's opinion.
You're not exactly setting a high bar for an LLM to overcome.
Dats Da Joke.
I've noticed that easily half of the County-level Judges I have worked in front of, especially those that have held their seat a long time without getting called up to Circuit level, are basically glorified clerks for all the legal reasoning they can do. They oversee an assembly line where parties are being shuffled along towards a particular outcome and the Judge just pulls the lever that rubber-stamps the outcome as 'legal.'
There's some selection effect, if you were making bank in private practice no way you'd accept a Judgeship with such little power. But yeah, letting County Judges use LLMs from the bench could only improve things.
Of course, if you ever ask me to identify which half of the Judges I'm talking about, I'll clam up because those are ALSO the ones most likely to be petty and make my job more miserable.
I had to google it since I'm not familiar with FL judicial structure--your county-level judges would be magistrates where I am, and they don't have to be lawyers here. Having them use LLMs for help would probably be a shocking level of improvement. Even the felony-level judges here (what FL appears to call the Circuit judges) who do have to be lawyers would generally benefit from the assistance.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
You might be right. Hasn’t been my experience.
I hope you are wrong (if I’m right we both have jobs)
Hoping for the best (AI makes the practice of law more tolerable/less mentally tolling), preparing for the worst (forced to swap to a career that requires working with my hands).
Well, I’m a bit worried that if AI solves law, robotics will solve “working with…hands” when combined with AI.
Not wrong.
But there was a pretty convincing case against it put forth in here.
Quoth:
I'm taking a gamble that martial arts/self defense instructors will still be in demand because people will probably prefer to have an instructor that is human, and a human is probably still going to be best suited to demonstrate techniques and movements that another human is expected to learn, and will be a strictly superior training partner, too.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
As an aside, older is not better for doctors. It's a common enough belief, including inside the profession, but plenty of objective studies demonstrate that 30-40 year old clinicians are the best overall. At a certain point, cognitive inflexibility from old age, habit, and not keeping up with the latest updates can't be compensated for from experience alone.
(This doesn't mean older doctors are bad doctors, it just means they aren't the best anymore, all else being equal)
I think there's a component for this, but if pushed I wouldn't say it's the biggest factor. An ophthalmologist hasn't studied cardiology since med school, they might remember the general details and interactions when it comes to the drugs they prescribe, but they're still not a cardiologist.
Gun to my head, I'd say that the human doctors who outperform LLMs are still smarter, if not as well read (they can't boast a near encyclopedic knowledge of all of medicine like any half-decent LLM can). IQ matters, and some doctors are just that smart, while having the unfair advantage of richer interaction with a human patient. Plus LLMs don't have the same "scaffolding" or affordances, they can't just look or lay hands on a patient (though they can ingest pictures, that's still an extra step). I suspect the difference diminishes to a large extent when the doctors are given the exact same information as an LLM, say some kind of case overview and lab tests + imaging. GPT-4 was scoring at the 95th percentile level in the USMLE, and these days medical benchmarks are simply not good enough to compare between them (official, graded benchmarks, I'm sure you can make a few more-ad-hoc ones if you really try, though by "you" I mean a competent physician).
I definitely believe that younger doctors are more up-to-date in best practices and aren't full of old knowledge that has proven ineffective or even harmful.
But if you could hold other factors approximately equal, I'd still bet my life on the guy whose' seen 10,000 cases and performed a procedure 8000 times over someone who is merely younger but with 1/3 the experience.
Lindy rule and all that. If he's been successfully practicing for this long its proof positive he's done things right.
Yeah, I suspect that even if LLMs are a full standard deviation IQ higher than your average doctor, the massive disadvantage of only being able to reason from the data stream that the humans have intentionally supplied, and not go in and physically interact with the patient's body will hobble them in many cases. I also wonder if they are willing/able to notice when a patient is probably straight up lying to them.
And yet, they're finding ways to hook the machine up to real world sensor data which should narrow that advantage in practice.
And as you gestured at in your comment... you can very rapidly get second opinions by consulting other models. So now that brings us to the question of whether the combining the opinions of Claude AND Gemini AND ChatGpt would bring us even better results overall.
https://www.nature.com/articles/s41598-022-15275-7
I don't think 14% is a big deal, there's already a great deal of heterogeneity in terms of surgical outcomes for all surgeons overall, but it does exist.
While it's frustratingly hard to find actual sources on the average IQ of doctors, most claim an average of 120-130. IQ testing LLMs on human IQ tests like the Stanford-Binet or Weisler-IV is fraught, but I've seen figures around 130 from o3 onwards. If I had to wild-ass-guess, 130+ is a fair estimate for GPT 5.2T, and versions with enhanced reasoning budgets are making novel discoveries in mathematics, so...
(Did I say that IQ research in doctors is bad? Oh boy, just see what it's like for LLMs. There are papers still awaiting peer-review that use Claude 3.5 Sonnet. The field moves *fast".)
https://www.trackingai.org/home seems better than nothing and claims 140 IQ for GPT 5.2T on the public Mensa Norway test and 129 for the so-called offline version.
Make of that what you will.
I haven't specifically tested models on their ability to catch lies or inconsistencies, but I think they'd do okay, but probably worse than a decent human doctor. This is a moderate confidence claim, and could be ameliorated by giving them previous clinical records, but a video feed would be even better (doable today). I'm already zooted on stims and typing up massive comments instead of studying, or else I'd try it myself.
The Transformer architecture is a universal function approximator, LLMs are already multimodal despite the name, and in the worst case, they can ingest text instead of raw sensor data like humans often do. I don't look inside the SpO2 probe, I read the number.
I've seen some evidence that diversity of models is good, for models of similar general competence. Even so, just putting the same info into another instance of the same model is highly effective, and I wouldn't yell at someone who did that.
I'd also be suspicious that this could be an artifact of the older surgeons being handed the tougher cases, or handling older patients such that complications are somewhat more likely to arise.
Similar logic to that study about black babies getting 'worse' outcomes when treated by white doctors... which dissolves when accounting for the fact that the white doctors were getting the toughest cases of any given race.
At any rate I'm sure there's more direct ways to assess a surgeon's skills from the outside (although apparently just asking for their IQ results is out?) but finding one that's a reliable, hard-to-fake signal is the challenge.
The main thing I appreciate about LLMs is the general fact that I can ask them to detail the source of all their knowledge and they can generally cite and point to it all so I can double-check myself, whereas I'd guess most doctors would scoff if you tried to "undermine their credibility" in such a way.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link