The most prestigious law school admissions discussion board in the world.
|
 |
 
|
Chatgpt pro ($200/month) is 100% worth it in LAW
Poast new message in this thread
Date: August 13th, 2025 4:23 PM Author: Bipolar Circlehead All bluebook citations are accurate, filing is crisp, 40+ page startlingly intricate and accurate law. Prompt was three/four long paragraphs detailing precisely what I wanted--not some offhand find it shit from a partner, PLEASE note. I had a lot of familiarity with the subject.
But wow. Wow wow wow. 30 mins and it generated a masterpiece. Litigators who naysay it are either 1) poor prompt engineers (likely) 2) too cheap/poor to afford pro |
Date: August 13th, 2025 4:25 PM Author: Navy jap Ok nigger I'm going to try it but if it churns out slop i'm going to furk your pumo ass no condom (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182286) |
 |
Date: August 13th, 2025 4:26 PM Author: Bipolar Circlehead ONLY use pro research grade, only the $200 month one, and PLEASE give very detailed instructions (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182288) |
|
Date: August 13th, 2025 4:29 PM Author: Bipolar Circlehead AND also tell it how long you want response to be and be prepared to wait at least 30 mins if sufficiently complex. You can also paste in deposition transcripts etc. Pretty much unlimited space to paste ANYTHING you need to give it MORE CONTEXT (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182294) |
|
Date: August 13th, 2025 4:30 PM Author: Mewling bearded state faggotry HOW WILL IT PROTECT CLIENT CONFIDENCES??? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182302) |
|
Date: August 13th, 2025 4:32 PM Author: Bipolar Circlehead Yeah who gives a fuck at this point (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182306) |
|
Date: August 13th, 2025 8:14 PM Author: Green Fat Ankles Goyim 180 (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182920) |
|
Date: August 13th, 2025 9:27 PM Author: exhilarant hissy fit (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183168) |
|
Date: August 13th, 2025 4:33 PM Author: charcoal stage wagecucks never used it for law (not a lawyer) but its lol how 180 good it is in general. anyone here who isn't a lawyer or whatever should quit their job immediately get a subscription and use it to start a one man business. you can literally do shit that would take months and a team in like days. i used it to design an experiment in pytorch that I ran for a few days and it generated a folder of like 100 graphs and dozens of parquet and json files worth of data. It took me like maybe 2 hours to refine a script that did all of that, probably would have taken weeks to write and debut it last year (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182310) |
|
Date: August 13th, 2025 4:35 PM Author: aphrodisiac pisswyrm (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182315) |
|
Date: August 13th, 2025 4:36 PM Author: Mewling bearded state faggotry Please explain in a way that a lawyer can understand. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182316) |
|
Date: August 13th, 2025 4:59 PM Author: Big hateful library CHATGBT FILE THOSE BRIEFS FOR THE FUCKLAW CASE OR I WILL FUCKING KILL YOU NIGGER (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182370) |
|
Date: August 13th, 2025 5:29 PM Author: Bipolar Circlehead I know joking, but if you're absolutely scorched earth in criticism it beautifully handles those with precision and ease. Much better than handlng to some associate you have to treat with kid gloves (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182472) |
|
Date: August 13th, 2025 7:52 PM Author: fear-inspiring philosopher-king I told it to make a roguelike farming sim metroidvania dating game and it did everything and the game is already in the top 20 on Steam selling like hotcakes (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182848) |
|
Date: August 13th, 2025 8:09 PM Author: charcoal stage wagecucks holy shit lol 180 if true (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182910) |
|
Date: August 13th, 2025 4:37 PM Author: aphrodisiac pisswyrm yeah if you're doing anything real you gotta use a reasoning model and have it think a long time
you should basically write the prompt like an email instruction to an assistant, too. give it as much context as possible. you can just ask it how to best prompt it btw. in general the best way to figure out the best way to get what you want from AI is to ask the AI |
|
Date: August 13th, 2025 4:47 PM Author: Bipolar Circlehead (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182347) |
|
Date: August 13th, 2025 8:28 PM Author: charcoal stage wagecucks yeah this is exactly what I do (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182950) |
|
Date: August 13th, 2025 4:37 PM Author: appetizing cruise ship Aren’t all Litigators Poor? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182322) |
|
Date: August 13th, 2025 4:48 PM Author: Bipolar Circlehead Some work in biglaw, friend (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182350) |
|
Date: August 13th, 2025 4:49 PM Author: Histrionic Base I/my office write maybe 5 briefs a year. Basic car wreck/summary judgment shit. Worth it?
|
|
Date: August 13th, 2025 4:56 PM Author: Bipolar Circlehead Yes (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182366) |
|
Date: August 13th, 2025 5:00 PM Author: Histrionic Base 180. On it. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182373) |
|
Date: August 13th, 2025 5:14 PM Author: Bipolar Circlehead o3 pro was 18000000 x's better than associates. I don't even care about them anymore (am SA) (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182418) |
|
Date: August 13th, 2025 8:11 PM Author: charcoal stage wagecucks I still use the o3pro api because its 180 and plus seems to work better when going cross model when you hot swap outputs (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182918) |
|
Date: August 13th, 2025 5:34 PM Author: beta chest-beating sanctuary karate You can just buy a month of it at a time for a case. Might even could bill to the client now that I think of it idk (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182502) |
|
Date: August 13th, 2025 5:43 PM Author: Bipolar Circlehead I've definitely used it for research in billing. Obviously you dont say used gpt but IMO it's saving clients a fuckload of time (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182529) |
|
Date: August 13th, 2025 4:51 PM Author: chartreuse lascivious point It's fine at raw analysis and legal research but still majorly sucks at writing anything from scratch (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182357) |
|
Date: August 13th, 2025 4:58 PM Author: Bipolar Circlehead *using thinking version ($20/month) rather than research grade intelligence ($200/month)*
Hey i do it all the time because thinking version takes WAY less time. |
|
Date: August 13th, 2025 5:15 PM Author: chartreuse lascivious point I use $200/month, and 4.5 is better at legal drafting than pro o3, (or o5). And all of them are much worse at drafting legal briefs, discovery etc. than Gemini Pro 2.5. Gemini is worse than Paxton.AI. None of them will cite check as good as Lexis Protege.
To fully mitigate any risk of malpractice or looking like an idiot when using AI to shitlaw you need to use at least 5 different AI platforms simultaneously IME. |
|
Date: August 13th, 2025 5:00 PM Author: Internet-worthy vibrant corn cake I've found the opposite; the vast majority of its current utility is in the standard searches that you can get for $20/mo. (To be clear: (a) no fucking lawyer is too poor to pay $200/mo. version; and (b) everyone should in fact get pro, for several reasons, probably first among them that you'll develop the best user skills that keep you most abreast of the tech as it advances.) Pro still 100% does not find magic-bullet fact-to-fact cases that you still need to do 3 hours of drudge work for. It is still maddeningly lazy about refusing -- outright refusing! -- to pull **every** federal districts' jury plan to look for certain language and categorize certain commonalities ("here's a table showing 8 plans! let me know if you want me to keep looking!").
The REAL turning point on AI in law will be when openAI finally integrates chatgpt into your discovery file to crawl around and look for shit. (Facts win cases, not law.) I'm hoping to see this within a year. Right now you can feed it bite-sized pieces, and the results, honestly, are not very impressive (but mostly because the inputs are not very large). But when it starts being able to watch and transcribe 773 hours of video and read 1.2M pages of text and tell me where are the places where subordinates raised concerns about possible MRSA outbreaks, where do you start to see the beginning signs of an outbreak, and for each decedent, create a timeline of their symptoms and reporting -- THAT will be useful. THAT will allow me to let go some of my entitled spoiled rich-kid shitlib associates from CLS. |
|
Date: August 13th, 2025 5:04 PM Author: Passionate Insanely Creepy Theatre Twinkling Uncleanness the laziness is so fucking niggerish and weird
does it have good memory features compared to stadnard? like it will remeber al the past shit i put in well or i have to keep reminding it?
|
|
Date: August 13th, 2025 5:10 PM Author: Bipolar Circlehead Memory a million times better on pro
o3 pro remembered details i posted like 40+ questions ago like butter. And integrated them fully into future responses. Not used gpt 5 for convos that long yet |
|
Date: August 13th, 2025 5:11 PM Author: Internet-worthy vibrant corn cake I am actually really pro-AI; it's only in environments like xo -- with oodles of mid- to high-160s lsat scorers who believe strongly in futurism because people of their intellectual capacity were really supposed to be religious, but religion is dead -- where I'm constantly having to be like "lol there partner, let's not spaz your Emory hat off your head, or pee through your Michigan Law sweatpants over this."
No, to be honest I'm extremely unimpressed at the model's long-term memory. I tend to delete most of my threads after I'm done with them, but I also don't expect it to remember stuff from those threads; I don't know if this practice contributes to this issue. I have no idea why this happens and I do not view it as a 'core' criticism of AI (it also doesn't really bother me that much).
The laziness is so fucking bizarre it makes me want to snap sam altmans twiggy little fag arms off. Like half of what I want you to do is give me links to all ~50 fucking PDFs, goddamnit. |
|
Date: August 13th, 2025 5:17 PM Author: Bipolar Circlehead Used it for 2-3 years and it is infuriating at providing links, I'll give you that. But if you use lexis on other screen can literally lookup the cites and they're ALL accurate now.
Pro/research grade I'd guess has a different memory bank than the normal "thinking" gpt. Keep it in one convo and you can do miracles with pro |
|
Date: August 13th, 2025 5:20 PM Author: Internet-worthy vibrant corn cake I have always credited the Westlaw ai with not hallucinating in the way chatgpt likes to -- although, if you notice, you can straight-up tell it to give you case quotes so you can opportunistically sift through them, and it will 100% refuse to do so. I've always suspected that's WL's way of avoiding the issue. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182436) |
|
Date: August 13th, 2025 5:27 PM Author: Bipolar Circlehead Westlaw ai was better than lexis but is miles behind o3 pro and 5 research grade now, even with specific bluebook cites. There are no longer hallucinations using the $200 month model
I have no idea how openai coded it now but it's scary af.
Obviously I'm violating a million and one client confidentiality rules but if you dare to try it will be completely fucking shocked and convinced your career will soon be caput |
|
Date: August 13th, 2025 5:34 PM Author: Internet-worthy vibrant corn cake I could give two shits about client confidentiality in this context and have no idea why it's become such a hot topic. I blame the proliferation of "ethics" CLEs needing some topic to harp on endlessly about.
OpenAI says they don't train on user inputs, right? So why is this any different from any other tool, like WL or whatever my investigator uses? |
|
Date: August 13th, 2025 5:38 PM Author: Bipolar Circlehead Thank you dude, my thoughts too
While it is admittedly giving a lot of info to gpt, do you know how troublesome that discovery request would be for the other party to sit and wade through without AI? Oh wait then you could just send them a discovery request for THEIR gpt history. So there's an impasse, no?
I think gpt will quickly become like westlaw or lexis w/r/t discovery or bullshit like that. Just not realistic |
|
Date: August 13th, 2025 11:35 PM Author: nubile whorehouse sound barrier I’m persuaded that the people who think the AI churns out solid briefs are true rerards. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183422) |
|
Date: August 13th, 2025 5:05 PM Author: Navy jap Have you tried notebook LM? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182388) |
|
Date: August 13th, 2025 5:07 PM Author: Passionate Insanely Creepy Theatre Twinkling Uncleanness is it cr?
a lot of people liek it |
|
Date: August 13th, 2025 5:16 PM Author: Internet-worthy vibrant corn cake No. Honestly I haven't heard of it.
My big AI limitation (and yes, it's a huge one; I get it) is that I'm all in on ChatGPT as a product; I no longer do anything with claude, grok, etc. It was a deliberate choice and the kind of tradeoff we all have to make. It seemed, and still seems, like chatgpt is the accepted industry leader. |
|
Date: August 13th, 2025 5:55 PM Author: Bipolar Circlehead (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182556) |
|
Date: August 13th, 2025 5:31 PM Author: beta chest-beating sanctuary karate (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182486) |
|
Date: August 13th, 2025 5:07 PM Author: Mildly Autistic Chrome Parlor is chatgpt black? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182395) |
|
Date: August 13th, 2025 5:08 PM Author: Bipolar Circlehead If you paste discovery files in gpt 5 reaearch grade and give it a good outline i maintain it is way better than a first or second year associate at pinpointing exact lines in deposition that will turn the tide
You do mass torts? Yeah that might reach the end of what gpt can do. But it's still a hell of a lot better than giving to a clueless associate |
|
Date: August 13th, 2025 5:50 PM Author: Internet-worthy vibrant corn cake This I agree with. There's still no substitute for having done the depo yourself and tracked the good admissions in real time, but on those cases where you're coming in during MSJ or the leadup to trial I've had better luck with chatgpt updating me on depos than the depo-takers themselves -- even not considering the time saved by the associate not having to update you. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182543) |
|
Date: August 13th, 2025 5:19 PM Author: chartreuse lascivious point "The REAL turning point on AI in law will be when openAI finally integrates chatgpt into your discovery file to crawl around and look for shit"
They turned this on like 2 weeks ago
I made a huge thread about it and have been using it daily to do doc reviews in the background
It went through 1,000 pages of discovery the other day while I was doing other stuff and sifted out 5 key docs for my mediation brief that would've taken me 10-20 hours to find |
|
Date: August 13th, 2025 5:22 PM Author: Internet-worthy vibrant corn cake What? It told me like 2 days ago that it still can't do that, and you still have to upload file by file (-ish, obviously). (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182440) |
|
Date: August 13th, 2025 5:23 PM Author: Bipolar Circlehead Become a better prompt engineer, i.e. an actual LAWYER with QUESTIONS (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182448) |
|
Date: August 13th, 2025 5:28 PM Author: Internet-worthy vibrant corn cake Easy there, queer; I'm not the one taking "10-20 hours" to review "1,000 pages of discovery" for hot docs. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182467) |
|
Date: August 13th, 2025 5:22 PM Author: Bipolar Circlehead 180 and absolutely correct.
o3 pro also did that starting ~3 months ago too |
|
Date: August 13th, 2025 5:27 PM Author: henna irate plaza knife Is 1,000 pages a lot of discovery to you? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182459) |
|
Date: August 13th, 2025 5:30 PM Author: Bipolar Circlehead I would be fascinated to see if you could actually use it for one of your mass tort claims.
I bet you could, easy |
|
Date: August 13th, 2025 8:23 PM Author: henna irate plaza knife We don't do mass torts, but this thread has been eye opening and I really like it here. We're likely to get 50,000 emails from Frito Lay in a couple months, at which point I'll start a thread to ask people to explain to me like I'm a Boomer how I can use "AI" to find the juicy ones. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182937) |
|
Date: August 13th, 2025 8:32 PM Author: chartreuse lascivious point I'll enjoy never responding to your thread so you can keep flexing about reviewing documents yourself like a dumbass
|
|
Date: August 13th, 2025 11:56 PM Author: abusive french chef principal's office I'll help you EPAH, I was telling these proles about Gemini 2 years ago (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183456) |
|
Date: August 13th, 2025 8:03 PM Author: Electric fluffy boistinker hell What are you using? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182885) |
|
Date: August 13th, 2025 9:10 PM Author: Submissive blathering kitchen genital piercing how the fuck do you do this????? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183109) |
|
Date: August 13th, 2025 11:29 PM Author: nubile whorehouse sound barrier Why would it take you 20 hours to review 1000 pages in discovery? Retarded? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183414) |
|
Date: August 13th, 2025 11:48 PM Author: chartreuse lascivious point 1) How the FUCK would I know? 2) Why are you PROUD of being an expert on this?
LOL |
|
Date: August 14th, 2025 12:15 AM Author: abusive french chef principal's office How can you know the computer is doing a good job if you're not manually verifying the results? The day I get AI output that's more than 90% satisfactory I'll scream about it in all caps on the front page of xo for 96v hours uninterrupted (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183484) |
|
Date: August 14th, 2025 7:46 AM Author: Navy jap Why would kenny need to do a good job on a case where no one was injured and everyone involved was an undocumented migrant? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183770) |
|
Date: August 14th, 2025 4:55 AM Author: bateful filthpig (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183682) |
|
Date: August 13th, 2025 5:17 PM Author: pontificating hall deer antler Actually cld do more for shitigator solos than biglol, if I was still practicing I cld just have it do all my discovery, respond to BS motions, draft complaints and few times I have to deal with msj or mtd it wld prepare something 10x better. U cld have one man shops going up against the prestigious skaden and dla pipers of the world (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182426) |
|
Date: August 13th, 2025 5:21 PM Author: Bipolar Circlehead That is precisely correct and absolutely 180
I literally used it against biglaw firms and me + partner have beat them the fuck out. I never tell partner where I get these flawless briefs from--and admittedly I ALWAYS check its references and do occasionally add bits and pieces. Literally same exact shit you would so as supervising attorney for a hotshot associate. GPT 5 I literally trust more than nearly all associates. It is pure insanity |
|
Date: August 13th, 2025 5:30 PM Author: aphrodisiac pisswyrm (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182478) |
|
Date: August 13th, 2025 6:04 PM Author: pontificating hall deer antler the amount of scholarly treats that are gonna appear in random shitlol briefs is gonna be lolzy, these kike partners in BIGLOL are not ready for it (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182586) |
|
Date: August 14th, 2025 4:06 AM Author: Excitant citrine bawdyhouse DeepSeek actually invented a quote that I have used twice now in trial briefs for restraining orders: “There is no duty to treat an eggshell psyche with kid gloves.”
That quote was it’s summary of case law on the eggshell psyche. |
|
Date: August 13th, 2025 5:42 PM Author: Internet-worthy vibrant corn cake **Sponsored DLA PIPER poast. Mr. Turdskin is a paid spokesman and brand ambassador.** (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182523) |
|
Date: August 13th, 2025 6:03 PM Author: pontificating hall deer antler Correction: Mr. Turdskin serves as the Chai Officewalla for DLA PIPER Mumbai. He is currently engaged to the Managing Partner, Mr. Dikshit Agarwal. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182583) |
|
Date: August 13th, 2025 11:58 PM Author: abusive french chef principal's office the NSAM story (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183460) |
|
Date: August 13th, 2025 5:27 PM Author: comical space Eye-opening thread. Really interesting to think what the workforce is going to look like in 5 years. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182461) |
|
Date: August 13th, 2025 5:30 PM Author: aphrodisiac pisswyrm it will look like a lot of indians unless we expel and/or kill them (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182481) |
|
Date: August 13th, 2025 6:04 PM Author: pontificating hall deer antler lol wtf mods (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182587) |
|
Date: August 13th, 2025 5:33 PM Author: Bipolar Circlehead Every lawstudent should be shitting their pants right now. If associate not absolutely eager to serve (which, let's be honest, none in last 5 years have been!) I immediately ghost them and start intensely using gpt on my other laptop. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182494) |
|
Date: August 13th, 2025 5:36 PM Author: Unholy office Looks like junior litigators are fuct but trannies appear to be safe (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182505) |
|
Date: August 13th, 2025 5:41 PM Author: Bipolar Circlehead It's even better for tranny, but a LOT MORE context is needed and at that point it's simpler to do on your own (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182518) |
|
Date: August 13th, 2025 6:05 PM Author: pontificating hall deer antler even better for trannies cause u can just upload a trannie pic to chatgpt and it will confirm whether its a real true trannie so u dont have to fear having sex with a gross biological female (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182589) |
|
Date: August 13th, 2025 8:04 PM Author: Electric fluffy boistinker hell One firm conviction I've come away with recently is that wordcel attorneys are ill equipped to perform the iterated, experimental prompting that using these tools successfully requires. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182892) |
|
Date: August 13th, 2025 9:02 PM Author: chartreuse lascivious point 90%+ of practicing attorneys like epah will never be capable of sophisticated prompting like the upper 99th percentile of AI shitlaw trailblazers on autoadmit
Their muscle memory and brains are too hard wired into doing rote doc review summaries + pdf tabbing + opening MS word to construct a brief from scratch using the same basic bitch boilerplate shit everyone has seen before
Litigators who don't have all that baggage and also have a knack for sophisticated cross-model prompting with customized backend/github plug ins and built in organic learning and refinement cycles will do better |
|
Date: August 13th, 2025 9:23 PM Author: Electric fluffy boistinker hell Hey, EPAH put some factual allegations I made up from Claude into an amended complaint and then told the court he consulted with experts prior to filing, so I give him credit for that. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183153) |
|
Date: August 13th, 2025 11:24 PM Author: nubile whorehouse sound barrier No, they won’t do better because they will outsource their thinking to these mediocre AI machines that suck at actually constructing nuanced and precise arguments based on the actual facts of a case. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183413) |
|
Date: August 13th, 2025 11:49 PM Author: chartreuse lascivious point You're one of the people I'm talking about (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183445) |
|
Date: August 13th, 2025 11:59 PM Author: nubile whorehouse sound barrier Your AI briefs are garbage. (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183462) |
|
Date: August 13th, 2025 11:58 PM Author: Navy jap Grok please draft an email to Jose to remind him that "everyone gets a shot" and that he has the worst case of facet syndrome I've ever seen in 4 years of scamming insurance companies (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183459) |
|
Date: August 14th, 2025 12:06 AM Author: Unholy office From Grok: Subject: Don’t Forget, Jose
Hey Jose,
Just a little nudge to remind you that "everyone gets a shot" — yeah, even you. Also, in my four years of dodging insurance company nonsense, your facet syndrome is hands-down the most ridiculous case I’ve ever seen. Try not to make it worse, alright?
Cheers, [Your Name] |
|
Date: August 14th, 2025 12:15 AM Author: Green Fat Ankles Goyim thats garbage and low effort (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183483) |
|
Date: August 13th, 2025 8:29 PM Author: hairraiser ladyboy (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49182954) |
|
Date: August 13th, 2025 11:00 PM Author: curious hilarious yarmulke (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183392) |
|
Date: August 13th, 2025 11:10 PM Author: aphrodisiac pisswyrm good poast, you've obviously actually used LLMs in practice
wanna loan me $5mm so i can post train some legal models and architectures? ill give you 30% return |
|
Date: August 13th, 2025 11:51 PM Author: chartreuse lascivious point A transactional firm shouldn't be subscribing to online chats
You need to build an LLM trained on your own doc library and every past client communication and internal training module, or pay someone to do that |
|
Date: August 14th, 2025 12:10 AM Author: olive haunting striped hyena orchestra pit I’m an tranny and our firm has done this.
It’s good but not perfect. I use it as a first cut for drafting specific clauses, to help me think through things, to get me conversant on a certain topic, etc. but I have yet to have it draft me anything that I could send out.
I’ve also used it to double-check my reading of agreements with respect to discrete questions. It’s again good but not great - it misses a lot of nuance. |
|
Date: August 14th, 2025 12:13 AM Author: abusive french chef principal's office All the same issues arise in litigation. You can use it churn out drafts, but you have to proof the shit out of them, and RAG will never be a substitute for an actual database because it inevitably fails to find everything you're looking for. Whoever said it's good for building timelines is right, but if you ask it for every email where someone said "fuck that bitch" you can't count on it to give you all of them. Of course this wouldn't be obvious unless you review all the discovery responses yourself and find responsive material the AI missed, but it sounds like a lot of biglolyers don't do that (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183481) |
|
Date: August 14th, 2025 6:40 AM Author: Cruel-hearted shimmering mental disorder Lol STFU faggot. You're draftings skill are sooooo impressive (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183724) |
|
Date: August 14th, 2025 6:52 AM Author: abusive french chef principal's office poast/moniker synergy (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183729) |
|
Date: August 14th, 2025 11:04 AM Author: Fighting hospital What about Harvey? (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49184324) |
|
Date: August 13th, 2025 11:08 PM Author: vigorous violet gaming laptop Everyone in here is a faggot (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183401) |
|
Date: August 13th, 2025 11:11 PM Author: bistre domesticated athletic conference (faggot) (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183403) |
|
Date: August 13th, 2025 11:44 PM Author: buck-toothed rigpig I'm gay (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183435) |
|
Date: August 14th, 2025 7:55 AM Author: impertinent swollen indian lodge Yeah if you’re a STUPID NIGGER (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49183779) |
|
Date: August 14th, 2025 10:58 AM Author: Soul-stirring station (OpenAI shareholder) (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49184299) |
|
Date: August 14th, 2025 4:00 PM Author: Costumed hot half-breed (http://www.autoadmit.com/thread.php?thread_id=5762076&forum_id=2,#49185384) |