> Instead, the researchers reckon that chatbots have one big advantage over mortal interlocutors: access to huge amounts of information that they can call upon instantaneously. Conspiracy theorists tend to come to arguments armed with intricate, seemingly fact-filled, defences, which their reality-based opponents lack the information to answer. The bots helped to even the playing field.
While this may be completely true, the idea that you’re going to use Chatgpt to argue against any factual claims is actually ridiculous, as Chatgpt is not in anyway capable of evaluating factual claims in the way doing actual research could. I suspect the reason that AI is so good at countering conspiracy theories is simply because they are such colossally stupid arguments that you don’t even need to be accurate to call out how inaccurate what they write are, and that most conspiracy theories run entirely on group think, so the same argument you hear now has likely been debunked a hundred fucking times in the training data.
WAGRAMWAGRAM on
Well, this only shows people really think chatbots “know” things
Mephistopolees on
Conspiracism is an endurance test against a firehose of horseshit that relies on wearing down the will to continually relitigate the same lies ad nauseum and find an unopposed platform to spread it. It also thrives on essentially antagonizing interlocutors into blowing up and using that against them.
It also generally predictable, heavily recycled, readily debunked TP, and similarly can be countered by someone who merely has a repository of counters.
AI has infinite patience, never loses its temper, and can break out a deluge of pre-packaged counters with no effort. While a conspiracy theorist wont be moved, if the counter is well-sourced, it could readily move gullible third parties observing the “debate”
StopHavingAnOpinion on
Can someone explain to me the recent hate against artificial intelligence? I genuinely don’t understand. It’s a potentially useful tool in dozens of sectors of society. Yes, it’s not always good, and you have the people trying to capitalise on it, but it’s not Crypto. It has and continues to have widespread uses in lots of places.
If it’s fear of job losses, I get it, but the pure *seethe* I see on reddit any time the topic comes up is confusing for something like this.
5 Comments
From the Economist:
> Instead, the researchers reckon that chatbots have one big advantage over mortal interlocutors: access to huge amounts of information that they can call upon instantaneously. Conspiracy theorists tend to come to arguments armed with intricate, seemingly fact-filled, defences, which their reality-based opponents lack the information to answer. The bots helped to even the playing field.
[Link to the original Science journal article](https://www.science.org/doi/10.1126/science.adq1814)
While this may be completely true, the idea that you’re going to use Chatgpt to argue against any factual claims is actually ridiculous, as Chatgpt is not in anyway capable of evaluating factual claims in the way doing actual research could. I suspect the reason that AI is so good at countering conspiracy theories is simply because they are such colossally stupid arguments that you don’t even need to be accurate to call out how inaccurate what they write are, and that most conspiracy theories run entirely on group think, so the same argument you hear now has likely been debunked a hundred fucking times in the training data.
Well, this only shows people really think chatbots “know” things
Conspiracism is an endurance test against a firehose of horseshit that relies on wearing down the will to continually relitigate the same lies ad nauseum and find an unopposed platform to spread it. It also thrives on essentially antagonizing interlocutors into blowing up and using that against them.
It also generally predictable, heavily recycled, readily debunked TP, and similarly can be countered by someone who merely has a repository of counters.
AI has infinite patience, never loses its temper, and can break out a deluge of pre-packaged counters with no effort. While a conspiracy theorist wont be moved, if the counter is well-sourced, it could readily move gullible third parties observing the “debate”
Can someone explain to me the recent hate against artificial intelligence? I genuinely don’t understand. It’s a potentially useful tool in dozens of sectors of society. Yes, it’s not always good, and you have the people trying to capitalise on it, but it’s not Crypto. It has and continues to have widespread uses in lots of places.
If it’s fear of job losses, I get it, but the pure *seethe* I see on reddit any time the topic comes up is confusing for something like this.