4
votes
Am I having a discussion with an AI? If not am I an idiot? A mean person?
Link information
This data is scraped automatically and may be incorrect.
- Title
- @Intrepid-figment comments on Small-Scale Question Sunday for April 23, 2023 - The Motte
- Authors
- @Intrepid-figment
- Published
- Apr 24 2023
- Word count
- 53 words
This, to me, is trolling. Whether that's good/bad/right/wrong is a personal judgement call I suppose.
Definitely not an AI.
Fair perspective. I will say, however, on the rare occasions that someone takes the same approach with me, I have found it very fruitful and even changed my mind about some things as a result. So, assuming I am doing it with kindness, or at least curiosity, there is a golden rule element at play.
Playing Devil's Advocate in a formal debate where everyone knows you're taking on that role and you have actually researched the position can be productive, but doing it online and spur of the moment can be incredibly detrimental to the community and everyone involved... especially when you don't let the other people know you're doing that, and are talking about controversial political topics, deeply held convictions, and issues that truly effect people's lives.
E.g. Playing devil's advocate for a homophobic/transphobic/racist/bigoted position against an LGBT person or other minority is downright cruel. We get enough of that shit in real life with real people who genuinely hold those abhorrent opinions. And it's exhausting, frustrating, and depressing enough having to deal with those people as it is. We don't need or want someone playing advocate online for them too.
p.s. TheMotte is full of people who genuinely do hold those abhorrent opinions, which is why I really think you should stop promoting them here. Seriously. Please stop linking to them here. Please.
I will honor your request, but my links are not at all intended as promotional. I find their community somewhat suspect, but am highly curious about who they are and how they operate. My request here is genuine.
Also, for the record, there is a line where things get personal that I wouldn’t cross. I would never pretend ti be any kind of discriminatory just for the sake if argument. But I might argue against electric cars even though I believe personally they are part of the solution.
Incidentally, one reason I engaged with this being was becaue at first, and fifth, glance, this contractualism he espouses and “human reason” based morality could very easily for the basis for things like brutal authoritarianism, fascism, slavery, conversion therapy.
Thank you.
And good to know you never go that far when playing devil's advocate, but unfortunately a lot of people online (esp trolls) do. It's one of the reasons it's often so hard to take people in good faith when discussing things online, and so hard to moderate online discussions.
I’ll add one more thing, just for the tildes community generally. When playing devils advocate, it’s usually because i have some strongly held beliefs about issues not being given due attention. For example, with electric cars, advocates seem to be ignorimg the right now social harms from batteries, both from raw material extraction and disposal. And so I will sometimes take a devils advocate approach to highlighting this issue.
I think the distinction here is that you're aren't playing devil's advocate, you're just adding information you believe is relevant to the conversation. Like you I'm glad we're phasing out internal combustion for EVs, but I also feel it's taking away from a larger discussion on investments in public transit and pedestrian/bike infrastructure.
There's more discussion on this issue over here.
Thanks, @FlippantGod ;)
I've seen AI generated comments on reddit due to subs I was in when ChatGPT launched and they are quite different.
Possibly this is why your argument went as such. Twice recently (once here) I have inadvertantly been a troll. I might be too quick to prescribe it, but I think you've made my mistakes.
I was inspired to reread Tildes code of conduct, and eventually found this, maybe it will help.
So no, I don't think you were having a discussion, and not with an AI. I don't think you are an idiot, or mean. But you might have demonstrated some of the qualities of a troll. Maybe reevaluate your behavior under that lens in a few days and see then if you reach a different conclusion.
That link is brilliant and I will be referring to it again.
Here, however, I was attempting a socratic discourse (asking questions to get to the truth of the matter), as opposed to stating an opposing position.
I bet it would have been better to ask permission first, or at least identify what I was doing.
If you want to have meaningful conversations like that on internet forums with strangers, you'll have to soften your tone. Significantly. All those things in high school writing class they said not to do? Do them. Passive voice, "Uh", "Hmm", "I think" - add all those. People always assume assume an implied mildly antagonistic tone to neutral text from strangers, especially in the context of a "debate".
With that, if you're always shifting your stance to the opposing stance of the other person, they're just going to assume that you're a contrarian troll. In "real life", physical gestures, vocal tone, other signs, and generally more willing to give people a benefit of a doubt (more investment?) mean that you can more easily get away with this.
With just text, you'll have to inject a significant amount of peaceful goodwill into your tone for the other person to continue on with the conversation. Well, rather, you need to turn it from an "argument" or a "debate" to a conversation.
Important perspective.
For the record, I don’t always take the opposite stance on every issue, but isn’t that the definition of debate kinds? Also themotte seems to hold itself out as a debate forum.
Sure, but in the end there's a difference between what we ideally aspire to and what our monkey brains can actually do. Emotions are very deeply set and it can be difficult for the logical centers to fully control innate emotional responses. It's not a failing per se, just a reality.
Debate itself is almost always a dead end. If you want to convince someone or something (or just have a more pleasant interaction), elevating it to a conversation is always the practical path forward.
From the famous Carnegie book,
"You can't win an argument. If you lose it, you lose it; and if you win it, you lose it."
I haven't read the whole conversation you linked, but I would say it is most likely there is a human behind those words. While AI language models have advanced (and are advancing) quite fast, what we've seen from OpenAI and GPT-4 and other companies/models couldn't be as advanced as this.
This a discussion I have engaged in over at themotte, which I found recently and posted about on Tildes here.
I occasionally like to take discussions like these all the way to the mat. To this being's great credit, they are by far more articulate, thoughtful and persistent than anyone I have engaged with previously.
As a note, I will often assume a "devil's advocate" role in these kinds of discussions even (sometimes especially) against positions I actually do hold. Although, in this case, I am highly skeptical of a claim that human reason alone can provide a satisfactory basis for morality. Even without bring god or spirituality into it, I think at minimum any reasonable being would have to include a chaos factor into morality.
Are my questions mean spirited or meaningful? I certainly intend primarily the latter, although I am trying to probe what appear to me to be deficiencies in this being's position (although at repeated points they merely reference others' works without explaining what I should take away from them, and at least once, their reference seems at odds with their conclusion).
I have tremendous respect for the folks here at tildes, would love to hear what yall have to say on the subject(s).
Edit: p.s., I think this response of theirs demonstrates 1)they are not an ai (but maybe not?) 2)their true face (or the true face of a majority of users in their training dataset?) 3) that I was able to probe a weakness in their position for which they had no reasonable response, and therefore resorted to cussing at me. Later they call me a troll, a "whinger," and of having poor reading comprehension.
An abstract philosophical argument can wander around indefinitely (in theory) but people’s patience is often limited and I think it’s better to avoid such wide-ranging discussions. But it sounds like you enjoyed it, and maybe the other participant did too at first?
But after changing subjects a lot, eventually the other participant got frustrated, perhaps because it seemed like it would never end. I’m not sure you can conclude all that much from that expression of frustration, other than the conversation is getting tedious for them.
It seems like if you want a good ending then you need to be looking for a way to bring the conversation to an end before it wears out someone’s patience. And that’s not going to happen when you resolve all your disagreements. (This never happens.) You need to be looking for a way to sum up before then. Explicitly asking if they want to keep going might be a good idea.
I’m not sure what a good, or any end, looks lime for that matter. Isn’t philosophy, like all inquiry, never-ending on purpose? It seems the only real end is perfect understanding and harmony, which appears to be unattainable, at least in this timeline.
What does a satisfying end look like to you in this context?
Yes, a perfect understanding is unattainable, so I don't think there's any point in trying? Just feeling understood at all counts as a success. Explicitly pointing out something you agree on can help.
If you can mutually agree that the discussion has gone on long enough while being willing to talk again some other time, I that counts as a good ending. If you find any points of agreement and can clarify what you disagree on, even better!