Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GLM 5.2 is nearly as accurate as a human book keeper

toot-books.com

Post reply on HN
Print view View original

121–130 of 131 posts

Re: GLM 5.2 is nearly as accurate as a human book keeper

#121

Post by __MatrixMan__ »

fragmede wrote: post #114
__MatrixMan__ wrote: post #113

Earlier quoted context omitted.

I'm not so jazzed on state-issued UBI myself. It makes the state feel like they're special and I think that leads to bad behavior. We should use money issued by the most trustworthy entity around, whatever it may be at the time, and the state should have to compete for that spot.

Corporations, government, or billionaires. Those are the three options, pick one.

There's also your friends and family and neighbors--anyone that you know in meatspace and maybe have a reason to trust explicitly.

Re: GLM 5.2 is nearly as accurate as a human book keeper

#122

Post by fragmede »

traverseda wrote: post #77
krupan wrote: post #60

Earlier quoted context omitted.

Note to self: traverseda doesn't have 2factor auth on his email and his LLM seems to have full access. Hmmm

The token is stored on a LUKS encrypted drive on kwallet. You'd have just as much luck getting my password from my firefox installs sqlite DB. This is also how email clients in general work. Also this isn't really an agent. At least not a long lived one. Each email or transaction gets it's own session, the llm can make a few tool calls but must emit json as the final result. Very very short context lengths, very pred…

Still, it's scary enough of a threat that I've got a separate laptop running ChromeOS with no AI shit on it that I use to access my banking.

Re: GLM 5.2 is nearly as accurate as a human book keeper

#124

Post by Haven880 »

Diogenesian wrote: post #31

This shouldn't be ignored in the discussion here: The job performed by the humans was broader than what was requested of the model in this benchmark: humans also had to find the relevant invoices (searching through mailboxes, or requesting them from providers) and reason through any circumstances which cannot be inferred from the bank feed and invoices/receipts on their own. In the benchmark these circumstances are p…

AI is a glorified calculator in this case. It frees up human to do other things. It shouldn't be viewed as human replacer.

Re: GLM 5.2 is nearly as accurate as a human book keeper

#127

Post by ValentineC »

benjiro29 wrote: post #57
zerobees wrote: post #26

This is a prime example of a problem space where accuracy matters, but it also matters who ultimately goes to prison. I'm going to go out on a limb and guess it's not the LLM. If you're acting in good faith and your accountant does something crazy or evil, your liability is limited to some extent. You may get a tax bill but you're probably not gonna end up behind bars. But if your LLM decides to do a little bit of ta…

> If you're acting in good faith and your accountant does something crazy or evil, your liability is limited to some extent. From my understanding, you are the person signing off on the paperwork that is submitted to the IRS. There is this cache 22 with taxes. You are responsible, but you outsource it to a accountant. Because you are not knowledgeable about the taxes. But you are expected to be knowledgeable to under…

> Whatever you submit for your company, you are again responsible. Even if you hired a accountant.

> So while technically, if a accountant makes gross mistakes, the bill will always fall in your lap, because you are expected to understand the reports you submit to the IRS. And catch any errors before doing so.

I've had a tiny accounting firm take on more and more accounting debt over years just to get the accounts (lots of journal entries into contra accounts; this was before I took over the company), that I was mostly rolling my eyes at how horrible they've become when I tried enlisting Claude's help to untangle them and hopefully move away from using that firm.

I know bookkeeping is tough, especially for newer businesses that haven't set up the processes, but a lot of the professional services industry feels like outsourcing a website to a freelancer on Upwork.

Re: GLM 5.2 is nearly as accurate as a human book keeper

#128

Post by josefrichter »

As someone who practised tax law for global companies for a few years, I can say that the correct classification of some invoices from the tax law perspective requires a full-blown legal analysis.

That involves full interpretation of the law, which includes "teleology" - that is, understanding the purpose and goals of the law. We are probably (?) not yet at the point where frontier models can do this better than humans.

In other words, it's the same question as when AI will replace judges. Not lawyers, but judges themselves.

Re: GLM 5.2 is nearly as accurate as a human book keeper

#129

Post by cudgy »

Haven880 wrote: post #124
Diogenesian wrote: post #31

This shouldn't be ignored in the discussion here: The job performed by the humans was broader than what was requested of the model in this benchmark: humans also had to find the relevant invoices (searching through mailboxes, or requesting them from providers) and reason through any circumstances which cannot be inferred from the bank feed and invoices/receipts on their own. In the benchmark these circumstances are p…

AI is a glorified calculator in this case. It frees up human to do other things. It shouldn't be viewed as human replacer.

Even when humans used to do the calculating?

Re: GLM 5.2 is nearly as accurate as a human book keeper

#130

Post by alexjimenez99 »

Calazon wrote: post #80
Diogenesian wrote: post #31

This shouldn't be ignored in the discussion here: The job performed by the humans was broader than what was requested of the model in this benchmark: humans also had to find the relevant invoices (searching through mailboxes, or requesting them from providers) and reason through any circumstances which cannot be inferred from the bank feed and invoices/receipts on their own. In the benchmark these circumstances are p…

My wife (head of accounting for a small business) has been working on automating large parts of her job using AI. It's not completely reliable and the human cannot be taken out of the loop, but the number of menial tasks she's been able to automate has been really cool. A lot of processing data that arrives in non-standard formats, generating documents based on that data, etc. She still has to review everything, but…

Curious what she automated? Invoices/receipt document processing been a round for a while. Did she go beyond that
Post reply on HN