The most prestigious law school admissions discussion board in the world.
|
 |
 
|
these models are superhuman geniuses at this point
Poast new message in this thread
Date: June 14th, 2026 10:38 AM Author: Cerebral boltzmann library its a fact. the burst IQ is untouchable by organic lifeforms (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937758) |
Date: June 14th, 2026 11:12 AM Author: Cerebral boltzmann library (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937829) |
|
Date: June 14th, 2026 11:12 AM Author: Confused turquoise pocket flask holy shit (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937830) |
|
Date: June 14th, 2026 11:19 AM Author: Impressive Home Blood Rage yeah i think a lot of people are in denial about how good at language tasks they are
they really are way better at writing than any human being at this point |
 |
Date: June 14th, 2026 11:20 AM Author: excitant frozen business firm electric furnace Nah writing is like their worst task the writing is still janky and filled with AI tells. The thing they way surpass humans on is logical reasoning, accuracy, knowledge, and processing speed. For instance maxing out an "lsat" is like an embarrassing retard toddler exercise for it. (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937852) |
|
Date: June 14th, 2026 11:21 AM Author: Cerebral boltzmann library its processing speed but in the context of an automated workflows and structured outputs tp (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937854) |
|
Date: June 14th, 2026 11:21 AM Author: Impressive Home Blood Rage they are extremely good at writing
just gotta prompt them correctly for whatever it is you want out of them |
|
Date: June 14th, 2026 11:23 AM Author: excitant frozen business firm electric furnace Well yeah, but you can't just lift the full output and use it. You have to curate the best sentences, continually prompt it to refine things etc. (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937858) |
|
Date: June 14th, 2026 11:24 AM Author: Cerebral boltzmann library wtf are you talking about. the actual agentic shit is an LLM outputting json that drives computer actions, db's and real time decision software (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937859) |
|
Date: June 14th, 2026 11:26 AM Author: excitant frozen business firm electric furnace I'm talking about the writing. But I agree that its totally cool that there are AGENTIC STRUCTURED WORKFLOWS PARSING STRUCTURED JSON THAT DRIVES REALTIME COMPUTER ACTIONS AND DATABASE FLOWS. One of the most exciting aspects of these systems (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937861) |
|
Date: June 14th, 2026 11:27 AM Author: Cerebral boltzmann library the writing is irrelevant and for loser corporate drones like lawyers and shit. automation is the tech frontier (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937862) |
|
Date: June 14th, 2026 11:36 AM Author: excitant frozen business firm electric furnace What about multi-agent debate with executable adjudication for lolyer jobs? Where you have a proposer agent generate a solution, an adversary tries to refute it, and then a judge calls tests, theorem provers, simulations etc. until finding a final answer grounded in executable checks. Seems like it could be useful. Mainlining was basically doing a proto-version of this with non-agent based systems (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937867) |
|
Date: June 14th, 2026 11:41 AM Author: Cerebral boltzmann library automating the paper pushing layer doesn't matter. let me know when robot judges adjudicate adversarial claims people didn't even need to "file." (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937872) |
|
Date: June 14th, 2026 11:47 AM Author: excitant frozen business firm electric furnace Yeah I agree that we need totally and completely ambient and pre-filing systems like this. A continuous institutional sensorium, with verifiers, procedural governors, a decision engine etc. no need to "file" anything at all (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937874) |
|
Date: June 14th, 2026 12:27 PM Author: Cerebral boltzmann library im telling you how the institutions will react to the influx of agentic lawsuits. my agent filed a defamation case against you after predicting your response to my second message above. (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937901) |
|
Date: June 14th, 2026 2:12 PM Author: excitant frozen business firm electric furnace 180 (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49938029) |
|
Date: June 14th, 2026 3:01 PM Author: Cerebral boltzmann library tell that to the model-as-judge, bucko (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49938071) |
|
Date: June 14th, 2026 4:04 PM Author: excitant frozen business firm electric furnace (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49938121) |
|
Date: June 14th, 2026 11:54 AM Author: saffron bisexual marketing idea crrrr (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937882) |
|
Date: June 14th, 2026 11:19 AM Author: excitant frozen business firm electric furnace The burst IQ for sure is so high it basically couldn't even be measured anymore by human instruments. But these things aren't running for 80 years like a human constantly processing new experience. (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937844) |
|
Date: June 14th, 2026 11:21 AM Author: Cerebral boltzmann library not yet (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49937856) |
|
Date: June 14th, 2026 3:06 PM Author: saffron bisexual marketing idea cr and embodied experiences might be more enlightening for them. but even without that, I wonder what would happen if, post-training, you had a chatbot read a bunch of novels one by one and reflect on each one afterward (interpret and judge it, come up with connections to what it's read before, etc.) and store its takeaways as part of its "personality" going forward (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49938079) |
|
Date: June 14th, 2026 3:07 PM Author: Cerebral boltzmann library ive talked to my AI's about this and they basically said that kind of thing is maybe effective in the short term but once they crack more efficient training that would move back down to the weights layer (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49938080) |
|
Date: June 14th, 2026 4:45 PM Author: saffron bisexual marketing idea hmmm interesting (http://www.autoadmit.com/thread.php?thread_id=5874160&forum_id=2],#49938244) |