The most prestigious law school admissions discussion board in the world.
|
 |
 
|
AI will destroy the world
| Taylor Swift is not a hobby she is a lifestyle | 09/15/26 |
| UhOh | 09/15/26 |
| Taylor Swift is not a hobby she is a lifestyle | 09/15/26 |
| Oh, you travel? | 09/15/26 |
| The Penis | 09/15/26 |
| Taylor Swift is not a hobby she is a lifestyle | 09/15/26 |
| average/ordinary/typical citizen/person | 09/15/26 |
| hank_scorpio | 09/16/26 |
Poast new message in this thread
Date: September 15th, 2026 10:03 PM Author: Taylor Swift is not a hobby she is a lifestyle (🇺🇸 🇵🇱) AI will not destroy the world by recursively improving large language models until they become some abstract superintelligence hellbent on human extermination. The more immediate threat is far more human, far more familiar, and far more plausible: a cartel of state power and private incumbents using fear to lock down the most important new means of producing and distributing knowledge.
That is why the encryption wars of the 1990s matter. Back then, the government treated encryption as a controlled capability rather than mere speech or mathematics. Once that move is accepted, the rest follows quickly: what looks like scientific publication becomes "export," what looks like open research becomes "proliferation," and what looks like competition becomes a "national security risk."
The same logic is now being aimed at AI. The public story is safety. The structural effect is permissioned innovation. If frontier models are placed behind licensing regimes, pre-release review, mandatory reporting, closed evaluation pipelines, heavy guardrails, and vague "duty of care" obligations, the result will not be the elimination of risk. The result will be that only the largest firms--the ones with Washington counsel, compliance departments, government relationships, and the capital to survive indefinite delay--will be allowed to build.
This is not just regulation in the ordinary sense. It is the potential conversion of a general-purpose intellectual tool into a regulated instrument that may be developed, released, or discussed only with official tolerance. Once that line is crossed, every alleged emergency becomes a justification for deeper control. Every failure becomes proof that the regime was not strict enough. Every independent actor becomes suspect by definition.
The ominous possibility is not that AI suddenly "wakes up" and kills us all. It's that AI panic becomes the predicate for a permanent legal architecture of surveillance, licensing, and prior restraint around computation itself. Not merely around harmful conduct, but around the creation and dissemination of knowledge. |
Date: September 15th, 2026 10:14 PM Author: UhOh bro we're all lawyers, you can just say regulatory capture. (http://www.autoadmit.com/thread.php?thread_id=5904162&forum_id=2�#50137225) |
 |
Date: September 15th, 2026 10:29 PM Author: Taylor Swift is not a hobby she is a lifestyle (🇺🇸 🇵🇱) this seems far worse than run-of-the-mill regulatory capture (http://www.autoadmit.com/thread.php?thread_id=5904162&forum_id=2в�€#50137237) |
|
Date: September 15th, 2026 10:29 PM Author: The Penis I think its funny that all of these researchers at big labs writing their sad little defectors letters about ai saftey are all announcing in the same breath that they got some high paying job at some new fancy safety lab for ai risk management. Another thing I'll say is that I think AI is dangerous but not because it is going to "break out" and start purposely causing catastrophic harm to humans. I think its dangerous because of how much it speeds up knowledge and research. The fact that a few million in compute can now brute force problems that took humanity a century or more to solve, and this is just the beginning, means we are going to reach new levels of speed and complexity to what we can solve and develop. Technological progress becomes unpredictable. (http://www.autoadmit.com/thread.php?thread_id=5904162&forum_id=2�#50137236) |
|
Date: September 15th, 2026 10:36 PM Author: Taylor Swift is not a hobby she is a lifestyle (🇺🇸 🇵🇱) AI is dangerous because LLMs are designed for content production, whereas social media and the rest of the modern Internet are designed for content consumption. Content production becomes a problem when normal people can genuinely threaten the elite's power. The elite want consumers, not producers of something better to consume. (http://www.autoadmit.com/thread.php?thread_id=5904162&forum_id=2в�€#50137239) |
|
Date: September 15th, 2026 11:46 PM Author: average/ordinary/typical citizen/person We will be under power rationing and then absolute unavailability long before this happens. AI is never advancing further than whatever it can do in next 6 months. (http://www.autoadmit.com/thread.php?thread_id=5904162&forum_id=2�#50137255) |
|
Date: September 16th, 2026 1:24 AM Author: hank_scorpio any retard with two brain cells to rub together always knew the threat of any tech is it being fully captured by the government and used to oppress, the future is a dystopia and this has been obvious for far longer than AI has been an even remotely real thing, our only hope is a full collapse, which on the bright side, seems pretty likely (http://www.autoadmit.com/thread.php?thread_id=5904162&forum_id=2�#50137301) |