The Ai thread

Ok, this is political in its content, so I have spoilered the screenshot. Let us just say

IMG_0096.jpeg

In case you are not sure, the person in question is apparently the '60s/'70s musician whose best known songs are probably Society's Child and At Seventeen.
 
I just had to fill out my 2025 “self evaluation.” New this year on the form was a requirement that I justify why I haven’t risked my bar card and sanity by using LLMs like I am supposed to.
 
I just had to fill out my 2025 “self evaluation.” New this year on the form was a requirement that I justify why I haven’t risked my bar card and sanity by using LLMs like I am supposed to.
Seriously? Last time I checked, which was ... yesterday I think the latest incident was ... lawyers are still getting their asses handed to them in court for doing that by extremely unamused judges. Was this your firm asking this or an outside evaluation?
 
I just had to fill out my 2025 “self evaluation.” New this year on the form was a requirement that I justify why I haven’t risked my bar card and sanity by using LLMs like I am supposed to.
Seriously? I'll say one thing about being of as old as I am is I can say "yep, I'm out". Had a buddy who was at his job for over 10 years have to re-interview with an AI bot to keep his position. How goddamned humiating is that?
 
Seriously? Last time I checked, which was ... yesterday I think the latest incident was ... lawyers are still getting their asses handed to them in court for doing that by extremely unamused judges. Was this your firm asking this or an outside evaluation?

Seriously? I'll say one thing about being of as old as I am is I can say "yep, I'm out". Had a buddy who was at his job for over 10 years have to re-interview with an AI bot to keep his position. How goddamned humiating is that?

Yep, this is the firm, itself, asking this. People who don’t understand LLMs overestimate what they can do.

My kid is taking an honors class where they create entrepreneurial projects to help society. As part of that they were forced to practice making smalltalk. They were told to do this by making smalltalk with an LLM. They also do practice job interviews - with an LLM.

it’s insane.
 
Yep, this is the firm, itself, asking this. People who don’t understand LLMs overestimate what they can do.

My kid is taking an honors class where they create entrepreneurial projects to help society. As part of that they were forced to practice making smalltalk. They were told to do this by making smalltalk with an LLM. They also do practice job interviews - with an LLM.

it’s insane.
I think there needs to be a happy medium somewhere because we know it's the way of the future, the problem is when it's wrong it is convinced that it's right anyway. I've asked specific questions related to drones and regulations that were patently wrong, yet stated as fact.

Putting blind faith into this is a really bad idea and if you have to fact check it then you're defeating the purpose.
 
honors class where they create entrepreneurial projects to help society. As part of that they were forced to practice making smalltalk. They were told to do this by making smalltalk with an LLM. They also do practice job interviews - with an LLM
Likely insane - it heavily depends on the quality of accompanying guidance by whoever is giving the class...
Like it or not, this will likely be a big part of the future, so learning how to deal with LLM output etc will become really important, and will need to be incorporated into the curriculums.

This will of course be highly counterproductive if the main goal is to offload effort from the teachers and "optimize" things.
 
Likely insane - it heavily depends on the quality of accompanying guidance by whoever is giving the class...
Like it or not, this will likely be a big part of the future, so learning how to deal with LLM output etc will become really important, and will need to be incorporated into the curriculums.

This will of course be highly counterproductive if the main goal is to offload effort from the teachers and "optimize" things.
What’s crazy is that this class has heavy parent-involvement. Many of the parents live in mansions and are founders or heirs of founders of giant companies we’ve all heard of, political action committees, etc. Even under-achieving me has interviewed and hired dozens of people. Everyone is willing to volunteer to mock interview the kids. But, nope. Some AI app knows better.
 
What’s crazy is that this class has heavy parent-involvement. Many of the parents live in mansions and are founders or heirs of founders of giant companies we’ve all heard of, political action committees, etc. Even under-achieving me has interviewed and hired dozens of people. Everyone is willing to volunteer to mock interview the kids. But, nope. Some AI app knows better.
There's so much more to an interview than just one's technical ability for the role, for example I knew it was easy enough for candidates to google answers (just before the AI craze) so instead of giving a simple "how do you build a search service application" I would ask them to walk me through the process and provide some examples. They didn't have to be 100% correct on every step but if they could give me the basics with common sense it went way further with me than someone who was rigidly booksmart. I also wanted to feel comfortable putting them in front of a client so it was far more nuanced. I can't imagine routing something like that through an AI bot, it's insulting.
 
"Why are you not using AI to assist you with your work?" Meanwhile, Google AI...

google-ai-knows-this-sub-is-the-best-source-of-information-v0-vuh84tns81ig1.png
 
"Why are you not using AI to assist you with your work?" Meanwhile, Google AI...

google-ai-knows-this-sub-is-the-best-source-of-information-v0-vuh84tns81ig1.png
Hey @Eric there seems to be a small visual/logic bug that's causing this new post to appear in the wrong order, above posts from yesterday, and the thread is marked unread even though I've read it. It's not super important, but I thought I'd bring it to your attention just in case of something more serious. And now my post is also in the wrong order too.

Screenshot 2026-02-07 at 8.42.39 AM.png
 
Likely insane - it heavily depends on the quality of accompanying guidance by whoever is giving the class...
Like it or not, this will likely be a big part of the future, so learning how to deal with LLM output etc will become really important, and will need to be incorporated into the curriculums.

That's what I am starting to do now. We all know that students are using AI all the time, so we can at least teach them to use it constructively.

Chatbots definitely improve my working efficiency, but one should not rely on them for drawing any kind of conclusions, as the quality is unpredictable. As a sparring partner/fill the gaps tool, they are invaluable for me.
 
Hey @Eric there seems to be a small visual/logic bug that's causing this new post to appear in the wrong order, above posts from yesterday, and the thread is marked unread even though I've read it. It's not super important, but I thought I'd bring it to your attention just in case of something more serious. And now my post is also in the wrong order too.

View attachment 38033
Wow that is odd, not sure what happened there. It's like it did a merge or something, okay let's see if it happens again.
 
Back
Top