Oh sweet baby Jesus no, burn AI to the ground, humanity can't be trusted with it
- Thread starter TheMysteriousGX
- Start date
Recommended Videos
Again, for whoever hasn't been paying attention: The purpose of a Large Language Model is to write words in an order that seems to make sense. It is built to tell the use what it calculates the user wants to hear. What is true is inconsequential; what matters is what "seems right", even the LLM has to invent what "seems right" out of whole cloth.
The people pushing AI don't care about the consequences, so long as they can make money off of it right now.
I like this take on AI art
The Guardian asks: Are we living in a golden age of stupidity?
www.theguardian.com
Seeing the part where people using computer assistance literally registered less brain activity on an EEG monitor was kind of chilling. Tools make our lives easier, but what if we start making things too easy?
Are we living in a golden age of stupidity?
From brain-rotting videos to AI creep, every technological advance seems to make it harder to work, remember, think and function independently …
www.theguardian.com
Seeing the part where people using computer assistance literally registered less brain activity on an EEG monitor was kind of chilling. Tools make our lives easier, but what if we start making things too easy?
That has been argued over all sorts of tools. For example, when GPS was first coming out for ordinary folks , people were always said to be driving off cliffs because the GPS said to.The Guardian asks: Are we living in a golden age of stupidity?
Are we living in a golden age of stupidity?
From brain-rotting videos to AI creep, every technological advance seems to make it harder to work, remember, think and function independently …
Seeing the part where people using computer assistance literally registered less brain activity on an EEG monitor was kind of chilling. Tools make our lives easier, but what if we start making things too easy?
How much of that is true I can't day, but it's been a concern for many a year.
Amazing. Like the stages of grief, the stages of a bullshit artist follow an observable arc when the bullshitter abandons one false claim to retreat to another, while marginally closer to the truth, still is self serving bullshit. This was exemplified most when people saw the trump and co admin spelll out the stages most clearly for even those asleep at the back of the classroom.
Anyway, seems the AI industry and its "leaders" are at the stage of "yeah ok it's a big dumb bubble of all spend no return lol, but that's a good thing!"
Anyway, seems the AI industry and its "leaders" are at the stage of "yeah ok it's a big dumb bubble of all spend no return lol, but that's a good thing!"
Maybe Frank Herbert was right. Time for a Butlerian Jihad.The Guardian asks: Are we living in a golden age of stupidity?
Are we living in a golden age of stupidity?
From brain-rotting videos to AI creep, every technological advance seems to make it harder to work, remember, think and function independently …
Seeing the part where people using computer assistance literally registered less brain activity on an EEG monitor was kind of chilling. Tools make our lives easier, but what if we start making things too easy?
If the last 30 years have taught me anything, it's that there is no limit to the human ability to believe things that they have reason to believe aren't true. Or perhaps it's more that their sense of trust isn't based on usefully verifiable reality.
For instance, you can have someone who complains that all the TV and the newspapers are a bunch of liars, and then for some reason they will let themselves be whipped into outrage by them. But why, if they don't think these sources are honest? I suspect because they aren't telling the truth when they say they don't believe the media. I think they tend to believe anything that it suits them to believe: which is to say, what conforms to what they already believe.
When they say they don't trust a media entity, what they mean is that this medium also publishes things that they disagree with. Or that "don't trust the media" is just a vacuous meme that people repeat because many others do, cynicism to pretend to be wise, without any meaningful appraisal of whether media is trustworthy.
Mm, $5 cost to produce per 10 seconds of video?
We should all pile onto there and put a massive dent in OpenAI's finances.
AI will protect our nation's children from school shootings bags of chips.
www.thebanner.com
Note that the police approached this 16-year-old kid with guns drawn because a shitty AI system couldn't tell the difference between Doritos and Derringers.
Baltimore County councilman calls for review of school gun-detection system
Julian Jones wants a review of the events that led to “police officers pulling up on a kid with guns drawn.”
www.thebanner.com
Note that the police approached this 16-year-old kid with guns drawn because a shitty AI system couldn't tell the difference between Doritos and Derringers.
AI doesn't need to be better at doing anything, it just needs to be cheaper.
The problem with AI is that it should be regulated, not stopped. And I don't mean the type of regulation that stops startups from competing against incumbents. Regulate the algorithms like Senator Elizabeth Warren suggested. But since the US is, at the moment, a majority gerontocracy, so really old people either want their pensions/stocks/retirement to keep going great, or they are ignorant of it. Likely the former for US leaders, since many of them have credible staffers who do know about reinforcement learning, AI, LLMs, automation, and AI workflows.
The second problem is the Putin problem of AI. The nation/person I am most afraid of getting AI isn't Iran, or can't get AI due to its lack of electricity(the last time I heard of Iran, they had an energy shortage), competent leadership, and general economic misery. O,r North Korea, which doesn't even have a functioning dense power grid based on satellite photos, despite lots of natural resources. It's not Xi who, if given AGI or nuclear-armed AGI, which is, despite what all leaders tell you, the eventual goal, would likely not nuke the US, given the fact that no one would trust them again, they have economic assets in Mexico, would face nuclear winter and possible large-scale starvation without US food, and fertilizers. And China can't just use AGI to disarm the US in terms of WMDs either, because our allies would give us a heads up, and they have their own AI.
The second problem is the Putin problem of AI. The nation/person I am most afraid of getting AI isn't Iran, or can't get AI due to its lack of electricity(the last time I heard of Iran, they had an energy shortage), competent leadership, and general economic misery. O,r North Korea, which doesn't even have a functioning dense power grid based on satellite photos, despite lots of natural resources. It's not Xi who, if given AGI or nuclear-armed AGI, which is, despite what all leaders tell you, the eventual goal, would likely not nuke the US, given the fact that no one would trust them again, they have economic assets in Mexico, would face nuclear winter and possible large-scale starvation without US food, and fertilizers. And China can't just use AGI to disarm the US in terms of WMDs either, because our allies would give us a heads up, and they have their own AI.
- It's Vladimir Putin. Putin can feed his people despite nuclear winter. Putin was the one who freaked everyone else out when he stated whoever controls AI will control the world, aka if I win the AGI race, I will actively use it against the West, whom he hates more than Xi or the CPC hates the US. I think the US and NATO have a strategic competition with China, not a Cold War(Yes, I was wrong), while the US has a Cold-Semi-hot War with Russia. If you let Putin be in control of AGI, he will use it to pinpoint the exact point every nuclear asset in NATO would be at its most vulnerable, and run historical models that show correlation with a black swan event and possible massive NATO nuclear vulnerability at its most critical point, and then he would strike.
- It's not China because China has not shown its willingness to go beyond the culmination point in war; they could have annexed more Indian land, they could have won the Sino-Vietnam War and taken land, but they have been very risk-averse. That's not to say they haven't thought the same way I am thinking, but their portability is lower, they are more in touch with the US-led order, as we speak, there is a clique of pro-west, pro-businessmen, pro-globalism clique in Shanghai, and likely in Guangzhou, Shenzhen, and other various coastal tier one cities minus Fujian.
Behold what the AI bros have declared will be "amazing games":
www.pcgamer.com
Personally, I love the part where the train car just slides into position over the platform. That's how trains work, right?
Sometimes I think this is how techbros see the world because they're just constantly blitzed out of their heads on ketamine.
Tech investor declares 'AI games are going to be amazing,' posts an AI-generated 'demo' of a god-awful shooter as proof
1v1 me at Canuial St.
www.pcgamer.com
Personally, I love the part where the train car just slides into position over the platform. That's how trains work, right?
Sometimes I think this is how techbros see the world because they're just constantly blitzed out of their heads on ketamine.
In all this doom and gloom over AI, it's important to highlight the success stories: AI can already perform on the same level as the average human US cop.
www.theguardian.com
US student handcuffed after AI system apparently mistook bag of chips for gun
Baltimore county high schools have gun detection system that alerts police if it sees what it deems suspicious
www.theguardian.com
Actually, I can think a way genAI can work: in a game that is deliberately dream-like. That trippy lack of persistence and logic might be a boon in that one specific scenario.Behold what the AI bros have declared will be "amazing games":
Tech investor declares 'AI games are going to be amazing,' posts an AI-generated 'demo' of a god-awful shooter as proof
1v1 me at Canuial St.
Personally, I love the part where the train car just slides into position over the platform. That's how trains work, right?
Sometimes I think this is how techbros see the world because they're just constantly blitzed out of their heads on ketamine.
Albania’s AI minister is ‘pregnant’ with 83 assistants: prime minister
Diella, the country’s first AI system that is now a minister in Edi Rama’s cabinet, will have 83 “children,” who will become parliamentary assistants.
www.euronews.com
