People have reputations. They cite sources. Unless they're compulsive liars, they don't tend to just make stuff up on the spot based on what will be probabilistically pleasing to you.
There are countless examples of ChatGPT not just making mistakes but making up "facts" entirely from whole cloth, not based on misunderstanding or bias or anything else, but simply because the math says it's the best way to complete a sentence.
Let's not use vacuous arguments to dismiss that very real concern.
Edit: As an aside, it somehow only now just occurred to me that LLM bullshit generation may actually be more insidious than the human-generated variety as LLMs are specifically trained to create language that's pleasing, which means it's going to try to make sure it sounds right, and therefore the misinformation may turn out to be more subtle and convincing...
The way in which this kind of error deviates from what a human would do is generally trivial: “confidently stating bs” is the same as how mistakes from human professionals often manifest—it will be this way anytime the person doesn’t realize they’re making a mistake.
The only real difference is that you’re imputing a particular kind of intention to the ai whereas the human’s intention can be assumed good in the above scenario. The BS vs unknowing falsehood distinction is purely intention based, a category error to attribute to an llm.
> The way in which this kind of error deviates from what a human would do is generally trivial
That's not even remotely true and if you've worked with these technologies at all you'd know that. For example, as I previously mentioned, humans don't typically make up complete fiction out of whole cloth and present it as fact unless those humans possess some sort of mental illness.
> The only real difference is that you’re imputing a particular kind of intention to the ai
No, in fact I'm imputing the precise opposite. These AIs have no intention because they have no comprehension or intelligence.
The result is that when they generate false information, it can be unexpected and unpredictable.
If I'm talking to a human I can make some reasonable inferences about what they might get wrong, where their biases lie, etc.
Machines fail in surprising, unexpected, and often subtle ways that make them difficult for humans to predict.
I don’t think you’re intending to impute intention, it’s just an implication of statements you made: “making stuff up on the spot” and “bullshit generation” vs unknowingly erring—these are all metaphors for human behaviors differing in their backing intention; your entire message changes when you use some form of “unknowingly erring“ instead, but then you lose the rhetorical effect and your argument becomes much weaker.
> that's not even remotely true and if you've worked with these technologies at all you'd know that
I have spent a good amount of time working with llms, but I’d suggest if you think humans don’t do the same thing you might spend some more time working with them ;)
If you try to you can find really bad edge cases, but otherwise wild deviations from truth in a otherwise sober conversation with eg chatgpt rarely occur. I’ve certainly seen it in older models, but actually I don’t think it’s come up once when working with chatgpt (I’m sure I could provoke it to do this but that kinda deflates the whole unpredictability point; but I’ll concede if I had no idea what I was doing I could also just accidentally run into this kind of scenario once in a while and not have the sense to verify)
> If I'm talking to a human I can make some reasonable inferences about what they might get wrong, where their biases lie, etc.
Actually with the right background knowledge you can do a pretty good job reasoning about these things for an llm, whereas you may be assuming you can do it better for humans in general than the reality of the situation
YouTube, Twitter, Facebook, newspapers, television, and auditoriums are filled with people that fill the world with pleasing sounding and utterly incorrect, or misleading, content. Humans are very good at convincing others their lies are true.
People don’t lie (“hallucinate”) in the way that LLMs do. If you’re having a friendly chat with a normal person they’re not going to start making up names and references for where they learned some fact they just made up.
Edit: Please stop playing devils advocate and pay attention to the words “in the way that LLMs do”. I really thought it would not be necessary to clarify that I know humans lie! LLMs lie in a different way. (When was the last time a person gave you a made up URL as a source?) Also I am replying to a conversation about a PhD talking about their preferred subject matter, not a regular person. An expert human in their preferred field is much more reliable than the LLMs we have today.
It's not about humans lying. It's about our memory getting corrupted over time where the stuff we think we're sure of is actually wrong or a misrepresentation of facts. Our recollection of things is a mix of real things and hallucinations. Witnesses provide wildly different accounts of the same event all the time.
This applies to PhDs as well and I don't agree that an expert human is automatically more reliable.
Are you sure about that? I can't count the number of times I've heard people spout marketing copy, word for word, to me while they think it's 100% true.
Are we talking about a conversation with a PhD in their preferred subject matter or not? That’s the line of argument I was responding to. I feel like as soon as we talk about LLMs the devils advocates come out of the woodwork.
They definitely do. I do all the time where I start explaining something just to realize that I'm actually not sure anymore but then it's often too late and the best I can do is add a disclaimer but most people don't.
Humans hallucinate all the time - first they consume propaganda/conspiracy theory and tell you lies while thinking they are right, and everybody else is wrong
A PhD will tell you if you're asking the wrong question. Human empathy allows us to intuit what a person's actual goals might be and provide a course correction.
For example, on Stack Overflow you'll see questions like how do I accomplish this thing, but the best answer is not directly solving that question. The expert was able to intuit that you don't actually want to do the thing you're trying to do. You should instead take some alternative approach.
Is there any chance that models like these are able to course correct a human in this way?
Jeesh, don't bring this up, you're apt to get ten people arguing about the X,Y problem instead, and why you should or shouldn't to 10 other things, rather than ask the user if they are on a legacy system where they can't make major changes.
It’s difficult to articulate but something as pure symbol of status as Mercedes is actually low status. Tesla, which is also an environmentalist symbol, is high status. This is an example of a pattern.
I would argue that Tesla is losing ground as a high status symbol, at least in the Bay Area. They certainly don't turn heads anymore, it's now Rivian, Lucid, Taycan, Ioniq 5, etc that are the talking points in the fancy grocery store parking lots.
What matters isn't the specific brand or even what they're selling, it's that it's The Cool New Thing.
The tech bro demo in SF loves high tech electrical stuff, and Lucid, Rivian, Taycan, etc. are, COOL and NEW and align to their political and social sensibilities, while also allowing them to flex the conspicuous consumption and signaling.
Mercedes is not new and does not align to obvious socio-political ideals outside of $$$$.
This is also mid-level bling in the grand scheme; the big money is still driving Lambos
I think that view might be some of the same generational (and cultural) mindset misunderstanding.
Fixations about cars (literally, or as a proxy for income), or about "our people", or "good families", are largely disappearing in the cohort of people who are starting careers and families today.
This trend started 60 years ago in the US, and is finally fully mainstream today, even if some parents are still holding on to the past (and if some communities are slower than others). But thank goodness for the progress, at least!
...
But the super-out of touch part of "new Mercedes" is that, even among the status/materially-motivated members of younger generations, a new car, and certainly a new Mercedes, ain't it.
"We want people who want a new Chevy! .. I mean, a new Lincoln! Cadillac? Camaro? Lotus? Mercedes? BMW TT? Prius?! Tesla?? Rivian??? Oh Jeez stop making it so difficult for me to sign up you kids to make me richer already!"
> This trend started 60 years ago in the US, and is finally fully mainstream today,
I assure you it’s not unique to desis in the US.
> even if some parents are still holding on to the past (and if some communities are slower than others). But thank goodness for the progress, at least!
It’s sad how many desis confuse “being more like white people” with “progress.”
Sure, I agree that it's not unique to the US. And I'm not trying to suggest that US/Anglo culture is superior in any way.
But I would suggest that children living their parents' lives, or for their parents' approval, is a recipe for deep dissatisfaction, unhappiness, and lost opportunities for many many people.
Some communities impose their "traditional" (usually hierarchical and patriarchal) expectations on the next generation more strongly than others.
And I would characterize moving away from those kinds of traditional expectations, as progress toward a more (difficult, but) chosen and intentional life. And I would call that a good thing.
> And I'm not trying to suggest that US/Anglo culture is superior in any way.
But that’s your ultimate point.
> But I would suggest that children living their parents' lives, or for their parents' approval, is a recipe for deep dissatisfaction, unhappiness, and lost opportunities for many many people. Some communities impose their "traditional" (usually hierarchical and patriarchal) expectations on the next generation more strongly than others.
In desi culture, it’s very important that children follow the right track, which their parents understand, instead of their own impulses. This cultural disposition is conducive to success—desi children are far more successful in American society than American children from similar economic backgrounds.
It’s also a recipe for contentment. You assert that living for what parents wants makes kids unhappy, but it seems to me that it’s American kids, who grow up being told to “follow their bliss,” who end up unhappy when their silly and unrealistic dreams fail to materialize. If you look at the statistics, from suicides to drug overdoses to cratering family formation, the western approach isn’t going that well.
> And I would characterize moving away from those kinds of traditional expectations, as progress toward a more (difficult, but) chosen and intentional life.
Individual “choice” is a distinctly western fixation. Worshiping the impulsiveness of youth over the judgment and experience of elders is a western fixation. It’s sad that so many desis conflate “being more westernized” with “progress.”
The "right track", you say. And parents know better. That's pretty self-serving, don't you think?
Self-determination and the ultimate responsibility of children to outgrow the default acceptance of their parents' belief systems are the absolute rights of every generation. Some communities acknowledge that fact, some encourage it, and some hide from it. Hiding won't make it go away.
Anyone who holds them back is committing an injustice against future generations.
It's not measured in economic success. That's a plainly myopic point of view. Freedom of activity and of thought is very noisy stuff, and it necessarily includes some failures. There are parallels to other systems of control exerted upon others here too, i.e. governments, etc. The seeds of those failures live in every community, they just are not dealt with as honestly in some.
The paternalist-authoritarian impulse to constrain the next generation's path is just old people not being able to admit that they don't know everything. This is like the landed gentry model of democracy. This way lies the death of the mind and the spirit.
Remember we're talking about overbearing parents having opinions about what brand of car their child's (inevitably, their daughter's) suitor drives. This is a trivial case of the parents' belief system failing the child. There are much more dramatic cases, of course, springing from the same self-serving impulses.
Financing is so readily available, that so long as you have stable employment, I think it is possible to procure most "luxury" cars, regardless of income. Whatever status symbol it may have conveyed as to wealth has been greatly diminished without going for the extreme upper end.
Spoken as someone who hates the necessity of owning a car. To me, it is a mode of transportation and that's it.
This is a false assumption, because some people definitely are. What is the real point of your statement? That you are not interested in opinions of such people?
Sure. You shouldn't care about the opinion of anyone who cares what car you drive, because they are, as made clear by that opinion, entirely superficial thinkers.
There's a comment higher from here that expensive car is a good proxy signal of wealth. Preferring wealthy people may feel unjust and even obnoxious, but it is definitely not superficial thinking. I personally would too steer clear (pun intended) of people like that, but just discounting them looks like the same behavior, isn't it?
At least where I live, what car you drive has mostly come decorrelated with wealth. Obviously at the extremes there’s still a relation (the guy with the beat-up 1990 civic is probably not rich, the girl with the $120k AMG wagon is), but is a new Tesla 3 driver richer than the person with a 2005 Subaru Outback? Who knows!
You might be thinking of counter-signalling on one hand. On the other hand there are just different groups of social status as well. To one group, a mercedes might be very impressive, to another it might be your time spent overseas, even if you all make the same money.
I think the reason cars might not be a reliable status symbol anymore is debt. If someons is in my socia-economic group but driving a very expensive car, it signals to me that they loaned the money so that they could have the status symbol.
Counter-signalling partly explains why pure status symbols are not among Gen Z. Taste, or aesthetic sensibility, is also an explanation. An expensive watch, such as a Patek Philippe, is still status since it's durable – to be passed on. A Mercedes? Not so much.
I also agree on your last point. Most young people today are unable to afford a home, or even pay off college debt. Wealth distribution is increasingly fat-tailed.
How is a Tesla an 'environmentalist symbol' when it has a 500kg battery with lithium mined from the global south and shipped around the world in the supply chain to add value?
Unlike the PhD, the AI model has benchmark scores on truthfulness. Right now, they're looking pretty good.