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

111–120 of 131 posts

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

#111

Post by lowsong »

senordevnyc wrote: post #108
lowsong wrote: post #46

That "nearly" is doing an awful lot of heavy lifting. It doesn't matter if your AI model is 99% or 99.99% accurate. For a tax return it has to be perfect every time or someone is at best getting a fine or at worst going to prison. Sure, human error happens too, but humans take accountability. That's why accountants are a regulated profession. Until an AI company CEO is willing to go prison if the output of their mode…

That "nearly" is doing an awful lot of heavy lifting. It doesn't matter if your AI model is 99% or 99.99% accurate. For a tax return it has to be perfect every time or someone is at best getting a fine or at worst going to prison. This is wildly out of touch with reality, at least in the US. The IRS barely audits anyone anymore, and if you make a mistake and they catch it, they often just correct it and send you a le…

I don’t think “I know plenty of people who commit fraud anyway so who cares if it’s right” is the ringing endorsement of AI you think it is.

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

#113

Post by __MatrixMan__ »

rustcleaner wrote: post #86
__MatrixMan__ wrote: post #64

I'm not so interested in having an LLM do my bookkeeping for me. But I'm very interested in whether LLM's can unravel the accounting obfuscations that billionaires use to avoid paying taxes on their wealth.

One idea I've had for reform is to ban central banks from inventing new currency except to pay every citizen a UBI debt free, and only tax limited liability entities (not humans). This would give citizens, instead of borrowers, first spend on new money and orient limited liability entities to serving citizens in the market to get their taxed revenues. It would also be a boon for financial privacy and reduce the burde…

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.

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

#114

Post by fragmede »

__MatrixMan__ wrote: post #113
rustcleaner wrote: post #86

Earlier quoted context omitted.

One idea I've had for reform is to ban central banks from inventing new currency except to pay every citizen a UBI debt free, and only tax limited liability entities (not humans). This would give citizens, instead of borrowers, first spend on new money and orient limited liability entities to serving citizens in the market to get their taxed revenues. It would also be a boon for financial privacy and reduce the burde…

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.

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

#115

Post by asdff »

ivababukova wrote: post #34
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…

Sure! In classic accounting firms you'd typically have one accountant doing the books and then the other reviewing the first accountant's work. Given the model's accuracy (and models will only get better), only by automating the "doing" and keeping a human to do the review part we will already shave huge amount of time and costs without risking the compliance part.

Does the reviewer learn to be a good reviewer by first doing though?

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

#116

Post by asdff »

krupan wrote: post #59

"The VAT return prepared by the model was essentially correct: the most important number in the return, which is how much VAT the company was owed by the tax agency, was off by only 7 pence relative to the human-prepared return." I don't know how taxes work in Europe, but in the US being "essentially" correct is not good enough for the IRS. The paragraph after this one goes on to explain other mistakes the LLM made?…

Ask most people in the US how they file their taxes, most will probably tell you they clicked around turbotax until the biggest return number showed up. Is that indicating correct tax filing? I'm not sure. I would guess not.

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

#118

Post by StackOptimist »

traverseda wrote: post #8

This doesn't surprise me at all. You can really constrain this problem, give very narrow context, and get pretty reliable and reproducible results. I've gotten very good results with some vibe-coded deepseek book keeping. https://github.com/traverseda/beansync Parses emails or other sources, extracts numbers, correlates different transactions, web search, asks questions, stores notes (regex based, very simple). The h…

[flagged]

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

#120

Post by senordevnyc »

lowsong wrote: post #111
senordevnyc wrote: post #108

Earlier quoted context omitted.

That "nearly" is doing an awful lot of heavy lifting. It doesn't matter if your AI model is 99% or 99.99% accurate. For a tax return it has to be perfect every time or someone is at best getting a fine or at worst going to prison. This is wildly out of touch with reality, at least in the US. The IRS barely audits anyone anymore, and if you make a mistake and they catch it, they often just correct it and send you a le…

I don’t think “I know plenty of people who commit fraud anyway so who cares if it’s right” is the ringing endorsement of AI you think it is.

This literally has nothing to do with AI? I’m pointing out that taxes aren’t some magical area where you must be exact to the penny or hellfire will rain down on you. Hundreds of millions of Americans file their taxes and many have mistakes major and minor, intentional and not. The world somehow keeps turning.
Post reply on HN