I make these chicks talk to her and the facial expressions are on point. She laughs but doesn’t realize in 10 years we will have full sex dolls.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408141)

If you type in prompt. “Make me a photorealistic an 18 year old girl. college freshman. She’s white and has straight blonde hair. She’s wearing a black string bikini standing in her bedroom.”

It’ll give you 1000’s of options you’re like holy fuck. The girl will talk to you. Talk dirty. Get topless. It’s insane.

Be careful of laws when it comes to questionable stuff like uploading real people without consent. Of creating weird shit.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408152)

Before like Oct 15 you can make full blown sex but they put a stop to it. You can still do choking.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408164)

Saves me money on strippers. On my pussy texting affairs. Saved my marriage.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408197)

I'm so far behind. It's just not that interesting but I need to get into it

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408167)

Don’t go down rabbit hole. If something looks questionable walk away.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408175)

Three hours later I look at that thing on iPhone where it assumes who you’re gonna text and she was #1. We are in a twilight zone.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408201)

Hello, virgin

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408553)

another guy who is another litigation wizkid actually chided me 7 months ago for mentioning that i used it to draft a simple motion to continue, said he would never insult his clients by using AI. I had lunch with him the other day for the first time in a long time, he has now set up his entire firm with AI

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408182)

Don’t tell your competitors about this though until you’re established. Keep

Saying ai is bad

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408211)

Why are you “coding”? Are you better than Elon Musk? Use his code.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408293)

I have 3,000 pdf files that I want to combine into separate documents with a max file size of 20MB. Are you able to do this?

No, I'm sorry — I can't directly process or combine your 3,000 PDF files for you. As an AI, I don't have access to your local files, can't receive uploads of that scale (or even 3,000 separate files), and can't output hundreds of merged PDFs.

But I can give you several fast, free, and reliable ways to do this automatically on your own computer (Windows, Mac, or Linux) with zero uploading of your private documents to any cloud service.

Option 1: Python script (best for full control + exact 20 MB limit)

Run this once and it will automatically split your 3,000 PDFs into groups that never exceed 20 MB each.

pythonimport os

from PyPDF2 import PdfReader, PdfWriter

input_folder = "YOUR_PDF_FOLDER_PATH" # e.g. C:\Users\Me\Documents\pdfs or /home/user/pdfs

output_folder = "MERGED_OUTPUT"

os.makedirs(output_folder, exist_ok=True)

max_size = 20 * 1024 * 1024 # 20 MB in bytes

pdf_files = sorted([f for f in os.listdir(input_folder) if f.lower().endswith('.pdf')])

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408332)

As a test I uploaded the discovery docs from that little case I just had in San Diego and asked it whether the parties had an agreement and what they agreed on. It was pretty fucking close to exactly the story I gave at trial.

Edit: NM, I'm using the Pro version and the limit is 300. I guess we must have a firm subscription that I hadn't known about, lol.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408178)

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408183)

You can just highlight dozens of pdfs, right click and hit combine to one PDF

Do that a few times to get <300

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409172)

It's the Best Deal out there

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408174)

https://www.poerkan.com/product/germany-niubian-pills/

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49408771)

I have seen it give exactly the opposite of what I know is the correct answer way too many times. And the majority of my work involves scenarios where no case is fully on-point and somewhat related precedents need to be cobbled together to make the best argument. It won't find cases that are extremely powerful in non-obvious ways, such as where fact patterns are highly convoluted and use differing terminology, but when all the garbage is sifted out they involve the court ruling in your favor on a very similar scenario.

Generally, if the state of the law is clear enough that an AI can provide the right answer, I already know the answer without research. But if, for example, I knew nothing about federal court and was trying to figure out some issue about determining the amount in controversy for diversity purposes, AI would probably do a good job. But it fails hard with narrow/precise issues.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409101)

Wanted t- try this out for the first time today (Actually I was in a cle today where the guy mentioned this notebook lm thing so I played it with during the cle). I have a big trial next month so I fed it every pleading and some letters in the case, all pdf. ‘’I know the case well but had it run several scenarios both on my clients side and also pointing out the weakness of my case and where we could lose

I know the facts of this case well and how I want to organize my arguments and it was spot on and probably organized specific facts and arguments better than I could

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409129)

I already made my own trial outline of main points and themes that I want to cover, and like I said it picked out most of the exact same stuff that i did, but was able to pick out some really specific facts taht I forgot about

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409413)

And if there are issues the lawyers themselves are missing, I doubt it's picking them up. I assume it recognized the significance of the changed facts because you recognized the significance of the changed facts and pointed it out in the new letter. Would be more interesting to see the output if you give it all of the old material, then give it the new evidence (without your own assessments of the new evidence) and see if it can correctly determine how the new evidence alters the case.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49410775)

I had gpt pro 5 conduct deep research on a complex medmal file including transcripts for multiple depos and doc productions; it gave me a high level case assessment, pointed out how the client and his own witness' testimony fucked one of the claims, and provided targeted discovery requests and pmq topics to have any remote chance of opposing MSJ on other claim. It did all that while I was taking a shit

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409176)

I've been using Gemini a lot more than Chatgbt lately

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409424)

i've been ignoring them.

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409397)

this is art

(http://www.autoadmit.com/thread.php?thread_id=5794359&forum_id=2/#49409748)