I’m curious how this increased throughput happens.

You’ve accurately stated that AI isn’t as rigorous as a trained attorney. Doesn’t that mean that every single datapoint must be confirmed by a human?

How is that quicker than just using a human to read the content and make the call? Data entry savings?

▲

ivraatiems a day ago | parent | next [-]

Yes, remember that these are effectively random PDFs in various different designs and formats, some of them not editable or even OCR'd.

It took a human attorney 20-30 minutes on average to manually copy-paste data from these PDFs into a spreadsheet (while also fixing any errors they found in the document and re-checking for quality).

Now, the AI copies everything into the spreadsheet in a small amount of time, and then the human reviews it. It takes maybe ~5-7 minutes to scroll to the appropriate pages in the document, read the lines vs the spreadsheet, and make corrections. So you've gone from 2-3 items an hour to ~8-10 items an hour.

Maybe you could pay someone to develop an OCR/ML application that could do this. But that project would never be profitable, even with the time savings. At the cost of a couple Claude subscriptions, it makes sense.

▲

newAccount2025 a day ago | parent | next [-]

Does the human find enough bugs that they stay on guard, or just rubber stamp everything without really looking at it? It’s hard to stay vigilant when stuff looks plausible.

▲

eru 21 hours ago | parent [-]

Complete speculation: you could instruct Claude to hide one random mistake in every document.

And Claude should write down the mistake in a sealed envelope, so it doesn't make into the database.

A review that doesn't find the mistake counts as invalid.

	▲	k4tsu 20 hours ago \| parent \| next [-]
		This is what bag scanners at airports do - the hit rate is so low and the job so boring the software projects fake contraband onto the imagery. Fail to spot the knuckledusters and expect a chat with the manager.
	▲	camdenreslink 10 hours ago \| parent \| prev [-]
		You could do this for code reviews. Have Claude include one subtle bug into each PR. If the reviewer does not spot it, then the PR is not mergable.

▲

ricky54 a day ago | parent | prev | next [-]

Don't forget that humans have a not insignificant error rate when copy/pasting or copy/typing data.

	▲	tkgally 20 hours ago \| parent [-]
		And it's possible to run each document through the LLM pipeline multiple times, using different models and/or prompts each time, to check for errors and inconsistencies. That will take more time and cost more, but it can reduce the error and hallucination rate significantly.

▲

barrenko a day ago | parent | prev | next [-]

I'm doing some public court records processing for bankruptcy cases (interested mostly to seek out corruption in big national cases), and yes, the "variousness" of random PDFs is exactly the issue. Trying to get the cost for a whole case down to a minimum.

Sample is around 300 court dates, shy under 1k files.

At best I'm building a claude skills file.

▲

stymaar a day ago | parent | prev | next [-]

> Maybe you could pay someone to develop an OCR/ML application that could do this. But that project would never be profitable, even with the time savings. At the cost of a couple Claude subscriptions, it makes sense.

A better use of these Claude subscription would be to develop the app (which it can pretty much do at that point) and you could iterate to make the workflow even more efficient than your current one.

▲

ivraatiems a day ago | parent [-]

Nobody working there has the requisite experience to do this in a reasonable amount of time. These are not particularly tech-savvy folks, Claude use aside.

	▲	stymaar 14 hours ago \| parent \| next [-]
		> These are not particularly tech-savvy folks, Claude use aside. The difference between a tech-savyy person, and a non-tech-savyy person has always been mostly in the later's head, but this is even more true now that we have pocket assistants who can answer pretty much all of our questions in a language tuned to our level of understanding.
	▲	eru 20 hours ago \| parent \| prev [-]
		Yes. And it might not even be worth it, as the AI agents gets cheaper and cheaper. Keep in mind that the task is fixed, so as the frontier of AI advances, you can switch to a cheaper trailing edge system and still get the same or even better performance for this task.

▲

refurb 18 hours ago | parent | prev [-]

Thanks for sharing the details. Does the attorney check that the AI copied the data accurately? Or is it just assumed to be correct?

Your experience mirrors my own. AI is great for parsing data that can take up a huge amount of time. My only concern is whether or not it’s done accurately. I wouldn’t use it for anything where mistakes cause serious consequences.

▲

juiceland a day ago | parent | prev | next [-]

You don’t need a trained attorney to schematize data. The LLMs are used to make the data easier to understand and manipulate.

▲

cromka a day ago | parent [-]

They'll also hallucinate and change meaning in the process of extraction and "schematization"

▲

margalabargala a day ago | parent | next [-]

Not necessarily. Depends how you use it.

"Write a python script that breaks down this PDF by X feature" would not hallucinate anything in the PDF. Certainly you could trivially double check that all text in the extracted JSON was in the text layer of the PDF.

▲

jeffbee a day ago | parent | next [-]

How much experience do you have with LLMs exactly? It would be consistent with my experience if Claude stuck in a line of python that just emits a JSON literal with no justification, potentially buried in a large program where an untrained person might not notice it. I don't even trust them if the output consists of structured data paired with source images from the PDF, because I've experienced LLMs fabricating the source rectangles to match the output. I only use tools like this by asking for programs, because as you note LLMs are good at that, and the verification process consists of tool calls to legitimate PDF manipulation tools so I have some confidence everything is above board. Even then I only do this for hobbies, not anything that matters.

▲

terminalcommand a day ago | parent [-]

Lawyer here. I used to trust Claude as hallucinations are near non-existent now. However for large volume tasks such as due diligence exercises, they still happen. We also tried Legora's tabular review, there were also numerous halucinated provisions in our due diligence exercise.

▲

egorfine 17 hours ago | parent | next [-]

Is it possible to catch those hallucinations using another LLM with a strong fact checker prompt with sources provided in output for human validation?

▲

rayiner a day ago | parent | prev [-]

Junior associates hallucinate too...

▲

NateEag a day ago | parent | next [-]

And when they do, you can train them or fire them, and they learn not to do it.

LLMs change not a whit, and there's no one to take responsibility for the failure (and thus no way to fix it).

As the new variation on the old theme has it, "A computer can never be held accountable, and so very many people are trying to get them make management decisions."

▲

IanCal a day ago | parent | next [-]

You can’t train people to never make a mistake, particularly when doing highly repetitive work like this. You must build your systems to account for that regardless.

▲

NateEag 15 hours ago | parent | next [-]

For sure.

But they do learn and improve.

The models don't (yet).

	▲	hollerith 15 hours ago \| parent [-]
		The models improve in the sense that GPT 5.6 succeeds at things GPT 5.5 fails at. It might be that the models have been improving in this sense faster than a human child improves.

▲

enraged_camel 15 hours ago | parent | prev [-]

Yes, exactly. Humans are non-deterministic as well, just in different ways. A tired human can make all sorts of errors for example, regardless of how much training they've had.

▲

juiceland 16 hours ago | parent | prev [-]

> LLMs change not a whit, and there's no one to take responsibility for the failure (and thus no way to fix it).

LLM output is nondeterministic and humans take responsibility for the failure the same way they take responsibility of a photocopy is too dark.

	▲	NateEag 15 hours ago \| parent [-]
		> humans take responsibility for the failure the same way they take responsibility of a photocopy is too dark. You mean, they notice it's too dark right after making it, change the settings, do it again, and give you the good copy? Because yes, that's my experience of humans.

▲

podocarp 21 hours ago | parent | prev [-]

You can scold juniors and they will learn. You can't scold Claude.

▲

stevesimmons 21 hours ago | parent | next [-]

Surely the rate of improvement in new LLM models is the equivalent mechanism?

▲

eru 20 hours ago | parent | prev [-]

You can scold Claude. Just doesn't make a difference.

	▲	chrisjj 20 hours ago \| parent [-]
		... until you hit Claude's risable "model welfare" protection.

▲

chrisjj 20 hours ago | parent | prev [-]

That's a recipe for disaster in my experience. I tried it (with Claude) on a simple tabular bank statement PDF, and it transposed two amounts, placinh each against the other's description. And the bot assured me the result was cotrect. The chance of a human checker catching such corruption is low.

▲

margalabargala 19 hours ago | parent [-]

Interesting. Did that PDF have a text layer or did you ask Claude to OCR it? If the latter I'm not surprised at all.

	▲	chrisjj 16 hours ago \| parent [-]
		Text layer.

▲

ferngodfather a day ago | parent | prev | next [-]

Doing similar-ish things with Claude, it's helpful to have something to ground it.

For instance, if you can say:

"Refer to the database schema in x.sql as your source of truth for the database structure we want to import into*. Do not invent data, tables or columns that do not exist. Carefully match all output against this database schema and do not create output that doesn't exist if it does not match the schema, simply skip it."

You will end up with a far better result in my experience.

Gotta treat it like a child.

	▲	chrisjj 20 hours ago \| parent [-]
		> Gotta treat it like a child. "Sorry for that, Your Honour, but we gave that case prep work to a child."

▲

ivraatiems a day ago | parent | prev | next [-]

They do, sometimes. That's why the review still has to happen.

But now it's comparing already filled columns on a spreadsheet, not copy-pasting every single thing from an (often uncopyable) PDF.

▲

chrisjj 20 hours ago | parent [-]

> But now it's comparing already filled columns on a spreadsheet

... with a PDF, right?

> not copy-pasting every single thing from an (often uncopyable) PDF

Obviously the PDF is copyable, else your bot would not be copying it.

	▲	ivraatiems 13 hours ago \| parent [-]
		"Copyable" in this case means you can highlight and copy paste text instead of manually retyping it.

▲

juiceland 16 hours ago | parent | prev [-]

But it gets it right like 99% of the time so human attention can be put towards catching the 1%, not entering data from one table to another and then catching that human’s mistakes.

▲

edmundsauto a day ago | parent | prev [-]

Reviewing something takes less time than producing it.

▲

lolakutty a day ago | parent [-]

I...don't think that is universally true.

	▲	edmundsauto a day ago \| parent \| next [-]
		Few things are. It also doesn’t require standards. “Stamp this diff” culture is everywhere even before AI. A stamp is literally easier than anything else. Whether that is useful measurement I suppose depends on the circumstances.
	▲	eru 20 hours ago \| parent \| prev [-]
		Yes, but empirically in this case, it seems to be true.