My first reaction was “who give a fuck?” then I got to the part of the article that says:
His website, which also features the purple dragon and a bunch of busted links in the footer, says that the firm “integrates AI to lower the cost of legal services.”
Which is honestly a thousand times more concerning than how he chooses to display his silly logo. Dude is writing legal documents with AI. At least his lack of professionalism is obvious.
I would imagine they’d be stupid if they did use AI. I’ve seen people use AI to “write” technical documentation that I have had to review. That shit goes straight into the bin because the time I spend fixing all the AI nonsense is about the same amount of time it would take for me to write the document myself. It’s gotten to a point where I straight up reject all AI generated documentation because I know fixing them is a waste of time.
I imagine legal documents have to be at least as precise as technical documents, so if they’re checking the output I seriously doubt they’re saving any time or money by using AI.
And anytime I see anyone advocating this crap it’s always because it gets the job done “faster”, and like, the rule is: “fast; cheap; good; pick two”, and this doesn’t break that rule.
Yeah, they get it done super fast, and super shitty. I’m yet to see anyone explain how an LLM gets the job done better, not even the most rabid apologists.
LLMs have zero fidelity, and information without fidelity is just noise. It is not good at doing information work. In fact, I don’t see how you get information with fidelity without a person in the loop, like on a fundamental, philosophical level I don’t think it’s possible. Fidelity requires truth, which requires meaning, and I don’t think you get a machine that understands meaning without AGI.
Then why don’t I just get the AI to write my legal filings and skip the lawyer fees?
The whole reason that I would retain professional help rather than just use an AI system is because the AI system makes stuff up, if the professional is just going to use the AI then what’s the point?
The problem with all of these types is they always say “oh well as long as you check its work it’s okay”, aka vibe<insert previously intellectually demanding task here>. The problem with that is firstly if you have to check its work then you’re basically just replacing one boring task with a different boring task, but you haven’t actually saved any time, and the second problem is a lot of the time people don’t actually check its work thoroughly, so mistakes still get through. Which is why using AI in a professional sense is a problem.
It’s fine to use it for summarising the minutes of a meeting, but it just isn’t good enough and reliable enough yet to be used in high-stakes situations. It is utterly irresponsible therefore to attempt to use it like that.
This is exactly the problem I have with programming tasks. It takes as long to check the code for problems (of which there are always many) as it would to write it and the code isn’t as good as mine anyway, and not infrequently just wholesale wrong.
For things like translating between languages it’s usually close, but still takes just as long to check as it would to do by hand.
Translating a code base between languages has always been a super easy, quick, fun experience for me. It’s like a simple exercise with clear right & wrong answers, a nice break from most coding tasks. Whenever I’ve done it I’ve felt like, “oh god, here goes the next week”, and then it’s done in a couple of hours max and nothing was stressful, and I feel great.
Sorry, machines can’t have that job. I do it better anyway.
My first reaction was “who give a fuck?” then I got to the part of the article that says:
Which is honestly a thousand times more concerning than how he chooses to display his silly logo. Dude is writing legal documents with AI. At least his lack of professionalism is obvious.
Pretty sure all law firms do that nowadays. And they are quite stupid if they don’t. As long as they check the output…
I would imagine they’d be stupid if they did use AI. I’ve seen people use AI to “write” technical documentation that I have had to review. That shit goes straight into the bin because the time I spend fixing all the AI nonsense is about the same amount of time it would take for me to write the document myself. It’s gotten to a point where I straight up reject all AI generated documentation because I know fixing them is a waste of time.
I imagine legal documents have to be at least as precise as technical documents, so if they’re checking the output I seriously doubt they’re saving any time or money by using AI.
And anytime I see anyone advocating this crap it’s always because it gets the job done “faster”, and like, the rule is: “fast; cheap; good; pick two”, and this doesn’t break that rule.
Yeah, they get it done super fast, and super shitty. I’m yet to see anyone explain how an LLM gets the job done better, not even the most rabid apologists.
LLMs have zero fidelity, and information without fidelity is just noise. It is not good at doing information work. In fact, I don’t see how you get information with fidelity without a person in the loop, like on a fundamental, philosophical level I don’t think it’s possible. Fidelity requires truth, which requires meaning, and I don’t think you get a machine that understands meaning without AGI.
Then why don’t I just get the AI to write my legal filings and skip the lawyer fees?
The whole reason that I would retain professional help rather than just use an AI system is because the AI system makes stuff up, if the professional is just going to use the AI then what’s the point?
The problem with all of these types is they always say “oh well as long as you check its work it’s okay”, aka vibe<insert previously intellectually demanding task here>. The problem with that is firstly if you have to check its work then you’re basically just replacing one boring task with a different boring task, but you haven’t actually saved any time, and the second problem is a lot of the time people don’t actually check its work thoroughly, so mistakes still get through. Which is why using AI in a professional sense is a problem.
It’s fine to use it for summarising the minutes of a meeting, but it just isn’t good enough and reliable enough yet to be used in high-stakes situations. It is utterly irresponsible therefore to attempt to use it like that.
This is exactly the problem I have with programming tasks. It takes as long to check the code for problems (of which there are always many) as it would to write it and the code isn’t as good as mine anyway, and not infrequently just wholesale wrong.
For things like translating between languages it’s usually close, but still takes just as long to check as it would to do by hand.
Translating a code base between languages has always been a super easy, quick, fun experience for me. It’s like a simple exercise with clear right & wrong answers, a nice break from most coding tasks. Whenever I’ve done it I’ve felt like, “oh god, here goes the next week”, and then it’s done in a couple of hours max and nothing was stressful, and I feel great.
Sorry, machines can’t have that job. I do it better anyway.
Pretty sure I’m not gonna hire you to do any professional work for me.
They were copy pasting most of the boilerplate before