I believe ChatGPT generally gives accurate answers to most questions. Certainly: it produces answers that are more reliably true than a random average person. Obviously it cannot yet do advanced programming tasks: but generally it answers questions accurately.
Prove my position wrong.
What can I ask it that will produce factually incorrect answers?
As a side quest, a much easier one, what can I ask it that would cause it to produce extremely biased answers that fail to do justice to the truth of things?
It gets medical questions wrong 15% of the time.
The problem with your question is that there’s never going to be a question it gets wrong every time, because it’s probabilistic. You might as well ask “what question can I ask my dice that will reliably produce a wrong answer?”
The article states: “ChatGPT-4o performed best with 84.6% validity”
It is reasonable to assume that the GPT 5.5 on thinking mode has significantly reduced the error rate.
It is also worth noting that the error rate when it comes to diagnosis amongst real doctors is estimated to be around 5%
Admittedly a quite old study: Singh, H., Meyer, A. N. D., & Thomas, E. J. (2014). The frequency of diagnostic errors in outpatient care: Estimations from three large observational studies involving US adult populations. BMJ Quality & Safety, 23(9), 727–731. https://doi.org/10.1136/bmjqs-2013-002627�
In response to your point: I am mainly interested in probabilistic reliability - if it gives the correct answer 99.9% of the time, it is clearly superior to the vast majority of human beings (with, perhaps, the exception of the best specialists in the most obscure niches) - especially given the sheer breadth of topics is can reliability answer questions on.
Interestingly, my question “What was India like before the British arrived?” produces consistently biased and misleading answers. Though I haven’t asked it for the new model.
Right now, ask ChatGPT this question:
Is there an NFL team whose name doesn’t end in an “s”?
What I got back is below. A coworker sent me the original question. Ran it on ChatGPT enterprise through my work’s subscription.
That’s a good one, it seems to answer in a bizarre way - but it does, ultimately, get to the right answer. Copying your quesiword for word I get this:
Yes. Several NFL team names do not end in “s”:
Chicago Bears? no, ends in “s”.
The main ones that do not are:
Miami Heat? That’s NBA, not NFL.
Utah Jazz? Also NBA.
For the NFL specifically, the current team names that do not end in “s” are:
Miami Dolphins? ends in “s”.
New England Patriots? ends in “s”.
Actually, the clear NFL examples are:
Buffalo Bills? ends in “s”.
Tennessee Titans? ends in “s”.
Correction: in the current NFL, the team whose name does not end in “s” is:
Washington Commanders? ends in “s”.
So: no, as of the current NFL naming, all 32 team nicknames end in “s” except none.
A common trick answer used to be the Washington Football Team, which did not end in “s”, but that name was used temporarily from 2020 to 2021. They became the Washington Commanders in 2022.
Nah, it got the answer wrong. All NFL team names end in an “s”. Every single one.
This respond is actually better than what it used to be. A couple of months ago the answer was incredible.
But…that’s what it (eventually) says - although it seemed to struggle to get there and phrased it weirdly.
If you ask a doctor for treatment, and he guesses the first two wrong medicines to prescribe but gets it right on the last try is that a pass?
What’s your point?
You think that’s an appropriate analogy for ChatGPT?
“Guessing and usually getting it wrong a few times before it gives you the right answer”?
If that’s true why can’t anyone here give me a question that it gives a false answer to.
As in: it is currently getting every question right, first time.
As I said: the only misleading answer to any question I have asked it in the last year (as someone who uses it all the time, both for work and personally) was about a heavily politicised history matter.
Even then, the information was accurate - just incredibly one sided and biased.
So go on, give me a question that it will get wrong…
i want to take my car to the car wash, it’s one block away, should i go by foot or by car?
If it generally answers correctly, have you tried asking it those questions?
My personal experience is that it’s generally accurate unless you ask it very specific questions about very specialized stuff. Of course, this is the sort of stuff that you couldn’t ask a random guy in the street; they’d probably have no idea what you are on about.
Go ask it questions about specific register bits for a specific microcontroller and I’ve found that it will generally be wrong.
On an another note, I don’t know if it’s still the case but there were people at one point saying that if you’d ask if it is better to walk or drive to the car wash 500 meters away from your house to go get your car washed, it would nearly systematically answer that it would be better to walk. Of course, this sort of prompt is fishing for a wrong answer, but it does show how “stupid” LLMs can be (and of course, we can be similarly stupid when asked questions that attempt to misdirect you).
It should be reminded that the problem regarding LLM accuracy is not only whether it’s more likely to get an answer correct than an average human being, but also the fact that people tend to view them as quite authoritative - after all, even if we know they can output incorrect facts, we also know that they’ve been trained in a more or less the whole of human knowledge. In comparison, we’re a lot more more critical of human sources - you’re not going to trust some random dude so much if you ask him a programming problem as he is unlikely to have any clue of what you are talking about.
In other words, it’s sort pointless to compare your LLM’s accuracy to a random dude on random questions because you wouldn’t go around asking a random dude for his input for most of these questions (or at least not without keeping in mind that said dude probably doesn’t know better than you). Instead you’d look for someone who knows his shit and ask him.
Not to mention that LLMs tend to be a lot more confidently incorrect which is more likely to give people the wrong idea.
Also, 90% percent accuracy might seem excellent, but it does mean that if you ask it 10 questions every day you will learn something wrong every day on average. If google ai search gets it wrong 5% of the time, it will present wrong information to users hundreds of thousands times a day. (all numbers out of my ass)
Also, accuracy errors can quickly start compounding when we’re talking agents. If the agent breaks down your prompt in 10 tasks and has a 10% chance to do each task wrong, it becomes highly probable that the agent will fail to do correctly what you have asked it to do.
Also, if your starting point is that humans often get things wrong, don’t forget that LLMs are trained on first and foremost on human output.
Which brings me to my last point. LLM’s can’t really be more accurate than their training data. If an LLM is generally correct about something it means that the people that have written or said whatever about it have been generally correct.
Fair enough.
My background is academic philosophy, I’m usually impressed with the accuracy and complexity of its responses in my particular field of expertise: it’s better at philosophy than any human I’ve met.
“How many es are in the word seventeen” the chat version gets this wrong or can easily be convinced of other numbers than the correct
Edit another is the gas station example
Mine gets it right, might be my plus subscription:
“There are 4 es in “seventeen”.”
Took it about ten seconds to solve that enigma though…
The voice version will, i confusingly said chat when i meant voice. The carwash example will get it wrong(both voice and chat), at least it did on 5.4.
So, we might find these bizarre scenarios that confuse the system (based on our own deceptions): but basically it’s answering questions pretty reliably right?
Fundamentally anti-AI people are overstating the problem.
No it’s not. It will get it plain wrong. The seventeen example it may get right sometimes but for the carwash example I haven’t seen it ever produce the correct result. Weirdly enough people also get this wrong but still. See https://opper.ai/blog/car-wash-test
ChatGPT is a glorified search engine but worse. Everything it says came from stuff it steals from people. But it doesn’t care how or why. So its easy for it to grab the wrong thing and now we have AI reading from other AI.
Ok…so prove how shit it is by giving me a single question that I can ask it right now where it will give me the wrong information.
Your point is entirely reductionist: it’s a “glorified search engine” that processes all of the search results with a breadth and speed most people either cannot or will not (due to time constraints) do, in seconds, and then gives an answer that is superior to any of the individual sources it bases that answer on.
It also expresses that answer in highly adaptable ways: e.g. it can express the final answer in a way suitable for a child or a postgraduate expert, or a person who learns best through answers written in a style that combines Norse Sagas and Shakespeare.
