The most prestigious law school admissions discussion board in the world.
|
 |
 
|
GPT-5 just going to be more expensive and do far less but businessfags need it
Poast new message in this thread
Date: August 6th, 2025 3:23 PM Author: Chocolate fantasy-prone stage It's going to be worse slop than ever before but this time the slop will have some necessary business function no one in business can live without. It's never going to 'get better' but they will figure out how to make more money by providing worse services which is what everything ultimately turns into. (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49162444) |
Date: August 6th, 2025 3:28 PM Author: Carnelian Step-uncle's House yup the only reason why this hasn't happened already, like it has in every other industry, is because AI is a brand new technology and so there is actual legitimate intra-industry competition
basically every single other "industry" in the modern world is a cabal with de facto price fixing and consumers have zero recourse except to pay the inflated rents |
 |
Date: August 6th, 2025 3:36 PM Author: Chocolate fantasy-prone stage Yep. We're just at the beginning of the cycle when everyone is trying to Get Gud. Give it 20 years. Enshittification is coming. In Phase I businesses love customers. In Phase II they hate suppliers. In Phase III platforms and suppliers hate customers and that's when it's all slop. (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49162500) |
|
Date: August 6th, 2025 3:43 PM Author: irradiated electric furnace dog poop please don't use that jew-contrived word. otherwise cr. (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49162515) |
|
Date: August 6th, 2025 3:51 PM Author: Chocolate fantasy-prone stage what would you call it (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49162552) |
|
Date: August 6th, 2025 6:14 PM Author: irradiated electric furnace dog poop you could even say "it will turn shit", "it will get shittier", "user cannibalization," pretty much whatever we like (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49162955) |
|
Date: August 6th, 2025 6:17 PM Author: provocative mexican trailer park (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49162958) |
|
Date: August 7th, 2025 6:01 PM Author: Judgmental flickering selfie candlestick maker only in the US dumbass, one reason china is DOMINATING is cause they believe in INSANE competition, not fraud jewish capitalism but REAL capitalism (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49165504) |
|
Date: August 6th, 2025 7:05 PM Author: Supple red native personal credit line Remember how much longer and more fluid the outputs were when o1 was first released? Then with o3 and o4 they got more "accurate" for coding but shorter and more robotic. We are already figuring out how to rip the goyim off. Much of it is pretending its all about "safety" and "alignment" while behind the scenes optimizing for shekel extraction (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49163052) |
|
Date: August 6th, 2025 7:22 PM Author: Chocolate fantasy-prone stage AI is never going to "do" anything the same way crypto is never going to be a thing in real life. AI is going to get massive private and public investments. It's going to be in the news every day. And BUSINESSES, businesses will insist that they use it all the time and that it's totally essential to whatever the hell it is that they supposedly do.
People, regular people like you and me, will never have a use case for it beyond what it can do right now. The parlor trick stuff will always be around and maybe it improves slightly. But there will never be any sensible use case for it for regular people. |
|
Date: August 6th, 2025 8:01 PM Author: Supple red native personal credit line This already isn't true. Top AI's are more intelligent than all but the top .1% of humans. Meaning a 145 IQ well educated human can now do the work of a 200 IQ human without having to hire the 200 IQ human (any lower than 140s and the human might be too dumb to even recognize how intelligent the AI is, and even if they do they won't know what to ask it to get worthwhile outputs). This makes a huge difference. (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49163180) |
|
Date: August 6th, 2025 8:16 PM Author: Chocolate fantasy-prone stage If AI isn't so smart what's one thing it's helped humans to do? Are there any real world examples of its performance? (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49163223) |
|
Date: August 6th, 2025 8:28 PM Author: Supple red native personal credit line I used it to write and optimize a quantum computing experiment yesterday that was borderline non human readable and basically required magic because of how shit error correction and instrument noise is. Might not feel real to you but it is to me. IDK much about the examples outside what I use it for, most of that shit is fake to control public perceptions and make them think its less capable than it really is (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49163272) |
|
Date: August 6th, 2025 9:28 PM Author: Chocolate fantasy-prone stage What type of useful real world computer coding will we be able to expect from it in the next decade. Major advances in software? (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49163395) |
|
Date: August 7th, 2025 12:47 PM Author: embarrassed to the bone cumskin Sounds like something that really happened (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49164633) |
|
Date: August 7th, 2025 1:52 PM Author: Chocolate fantasy-prone stage Every AI fanboy I know has told me they're on the verge of a major breakthrough in science and philosophy but I've yet to see the results. (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49164852) |
|
Date: August 7th, 2025 6:06 AM Author: dark razzle menage Before Deep Seek existed ChatGPT probably could have gotten away with doing that but now they can't. (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49163971) |
|
Date: August 7th, 2025 8:59 AM Author: Supple red native personal credit line Deep Seek sucks though (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49164108) |
|
Date: August 7th, 2025 10:45 AM Author: Chocolate fantasy-prone stage Cr it's really bad compared to Chatgpt (http://www.autoadmit.com/thread.php?thread_id=5759222&forum_id=2E#49164320) |