4 ms·

> as a meta point. I mean, it's a stretch. I could use that stretching metaphor as an excuse to talk about taffy for 4 paragraphs, but it doesn't contribute an

> as a meta point. I mean, it's a stretch. I could use that stretching metaphor as an excuse to talk about taffy for 4 paragraphs, but it doesn't contribute anything meaningful to the conversation. Scrounging up discussion about Web3 on an Indian government exposé is off-topic, and readers will flag it. It's equally substantial as those "this is good for Bitcoin" posts that get flagged under every article about a government treasury. > the topic is literally about how to authenticate things when you have an opposing central authority. If so, his flagged post didn't reflect it. It was a vapid comment that made no attempt to reconcile either topic. > his point was about how unpopular opinions are being proved correct If you want to argue that the Blockchain could have prevented this from happening, then by all means - write that comment. But that's not what his flagged post was. It detracted from the conversation and quite literally read as flamebait. > if you were downvoting/disagreeing with opinions about how AI is going to make it hard to authenticate information online, then you were wrong Spoilers: it has been hard to authenticate information online for as long as the internet has been around. In fact, the most significant events are corroborated with real life occurrences that challenge these lies. These women can attest that they did not do this in the photo. If government officials insist that the version with them smiling is the true master copy, then you're the fool for believing them. I'm baffled that this is even being used as a platform to discuss Web3, a collection of technologies mostly reputed for... being manipulated and exploited by bad actors that cannot be incentivized to stop.

stale2002 3y ago
> Scrounging up discussion about Web3 on an Indian government exposé is off-topic It really isn't. Because it is one of the normally very rare use cases for web 3 stuff, and the topic is directly about a case where such a thing would be needed. > it has been hard to authenticate information online That's nice and all. But the claim was that people were previously dismissing opinions about AI that are now mainstream. So yes, you'd be wrong if you were outright dismissing these opinions about AI before, which are now popular. And it is useful and interesting to point out how previous opinions that were dismissed out of hand, are now mainstream, using this situation as an example.
- smoldesu 3y ago
  > Because it is one of the normally very rare use cases for web 3 stuff > a case where such a thing would be needed. So... suggest a solution. What is your magical Blockchain panacea for this situation? To plug all digital cameras into an online ledger and hash every image? I'd love to hear a non-pipe-dream based idea for this. > But the claim was that people were previously dismissing opinions about AI that are now mainstream. This could be said about just about everything. I still dismiss the flying car, even though I was "proved wrong" in the 60s and 70s when some coked up engineers built one. I still dismiss cryptocurrency, even though many coins will outlive me. There is no world in my lifetime where cryptocurrency obsoletes the IMF. Betting against the powers that be is like playing chicken against a shotgun.
  - stale2002 3y ago
    > So... suggest a solution. I am not sure if you read the rest of the comments and suggestions in this whole post, but multiple people did, over multiple discussion threads. Feel free to go back and read them. Also, it is still interesting and useful to suggest part of the solution, without writing dozens of pages on the exact specific details of it. Even just recommending crypto, is interesting enough, due to the fact that real usecases for crypto are so rare. > This could be said about just about everything. Ok, well in this case, related to AI, lots of people were wrong. And those people who were wrong, should figure out what mistake they made, to be so wrong, and why they were so quick to incorrectly dismiss these arguments.
EGreg 3y ago
Somehow AI fans can’t seem to understand how scale changes things. Bombs have been around for a long time. What if everyone could have many, cheaply and easily? Why, since nothing happens with N of them, then 20,000 x N is just a little more of the same. And so what if the bots are now able to hold conversations and pass the Turing test 24x7 at scale? We have been able to hire humans to do that. Except… no. That’s not how it works and you know it. Allowing anyone to deploy swarms of bots with superhuman memory, strategy, coordination, and now the ability to do many tasks with acceptable or good results at scale, is a game-changer and the question is whether it will make society better off. The same questions asked of Web3’s downsides should be asked of AI. Not upsides, downsides. The fact that the guy in charge of safety at OpenAI says he thinks there’s a 20% chance of human extinction and you choose to shrug it off and focus on some other part — is the issue. This isn’t just an academic discussion. It is like the people who profit off of coal and other things, regardless of their emissions or effect on people or climate. It matters and should be regulated.

stale2002 3y ago

> Scrounging up discussion about Web3 on an Indian government exposé is off-topic It really isn't. Because it is one of the normally very rare use cases for web 3 stuff, and the topic is directly about a case where such a thing would be needed. > it has been hard to authenticate information online That's nice and all. But the claim was that people were previously dismissing opinions about AI that are now mainstream. So yes, you'd be wrong if you were outright dismissing these opinions about AI before, which are now popular. And it is useful and interesting to point out how previous opinions that were dismissed out of hand, are now mainstream, using this situation as an example.

smoldesu 3y ago
> Because it is one of the normally very rare use cases for web 3 stuff > a case where such a thing would be needed. So... suggest a solution. What is your magical Blockchain panacea for this situation? To plug all digital cameras into an online ledger and hash every image? I'd love to hear a non-pipe-dream based idea for this. > But the claim was that people were previously dismissing opinions about AI that are now mainstream. This could be said about just about everything. I still dismiss the flying car, even though I was "proved wrong" in the 60s and 70s when some coked up engineers built one. I still dismiss cryptocurrency, even though many coins will outlive me. There is no world in my lifetime where cryptocurrency obsoletes the IMF. Betting against the powers that be is like playing chicken against a shotgun.
- stale2002 3y ago
  > So... suggest a solution. I am not sure if you read the rest of the comments and suggestions in this whole post, but multiple people did, over multiple discussion threads. Feel free to go back and read them. Also, it is still interesting and useful to suggest part of the solution, without writing dozens of pages on the exact specific details of it. Even just recommending crypto, is interesting enough, due to the fact that real usecases for crypto are so rare. > This could be said about just about everything. Ok, well in this case, related to AI, lots of people were wrong. And those people who were wrong, should figure out what mistake they made, to be so wrong, and why they were so quick to incorrectly dismiss these arguments.

EGreg 3y ago

Somehow AI fans can’t seem to understand how scale changes things. Bombs have been around for a long time. What if everyone could have many, cheaply and easily? Why, since nothing happens with N of them, then 20,000 x N is just a little more of the same. And so what if the bots are now able to hold conversations and pass the Turing test 24x7 at scale? We have been able to hire humans to do that. Except… no. That’s not how it works and you know it. Allowing anyone to deploy swarms of bots with superhuman memory, strategy, coordination, and now the ability to do many tasks with acceptable or good results at scale, is a game-changer and the question is whether it will make society better off. The same questions asked of Web3’s downsides should be asked of AI. Not upsides, downsides. The fact that the guy in charge of safety at OpenAI says he thinks there’s a 20% chance of human extinction and you choose to shrug it off and focus on some other part — is the issue. This isn’t just an academic discussion. It is like the people who profit off of coal and other things, regardless of their emissions or effect on people or climate. It matters and should be regulated.