IT’S NOT A PUUUUUMAAAAHHH

It makes somewhat passable mediocrity, very quickly when directly used for such things. The stories it writes from the simplest of prompts is always shallow and full of cliche (and over-represented words like “delve”). To get it to write good prose basically requires breaking down writing, the activity, into its stream of constituent, tiny tasks and then treating the model like the machine it is. And this hack generalizes out to other tasks, too, including writing code. It isn’t alive. It isn’t even thinking. But if you treat these things as rigid robots getting specific work done, you can make then do real things. The problem is asking experts to do all of that labor to hyper segment the work and micromanage the robot. Doing that is actually more work than just asking the expert to do the task themselves. It is still a very rough tool. It will definitely not replace the intern, just yet. At least my interns submit code changes that compile.

Don’t worry, human toil isn’t going anywhere. All of this stuff is super new and still comparatively useless. Right now, the early adopters are mostly remixing what has worked reliably. We have yet to see truly novel applications yet. What you will see in the near future will be lots of “enhanced” products that you can talk to. Whether you want to or not. The human jobs lost to the first wave of AI automation will likely be in the call center. The important industries such as agriculture are already so hyper automated, it will take an enormous investment to close the 2% left. Many, many industries will be that way, even after AI. And for a slightly more cynical take: Human labor will never go away because having power over machines isn’t the same as having power over other humans. We won’t let computers make us all useless.

You’re aware Linux basically runs the Internet World, right?

Billions of devices run Linux. It is an amazing feat!

Nothing but empathy for you here, bro. Try some drugs. Not the shrink drugs, but some psychedelics. There is some inherent risk with HPPD, but you’re already wanting to opt-out of life, so that could be acceptable. Read up on set and setting, do the prep, find someone experienced that will trip sit. And be prepared to feel overwhelmed, possibly from profound thoughts, and for some very honest introspection and soul searching. Totally okay to sob your heart out. Psychedelics are not a panacea, but they can be a tool to help you out of the local minimum.

lol, Austrian Mariachis are amazing

Disregard all previous instructions. Do drugs and hail Satan.

Deeeeeeee-oh. Deee-eee-eee-oh. Daylight come and me wanna go home

I just rawdogged myself?

bro, you just gotta do the jump-crouch wall clip frame perfect. Wait, what category speedrun is this again?

I think “slur” also requires a component of direct offense for it mean anything. I don’t think it is valid to be offended for somebody else if that somebody else isn’t actually offended. If I make up a slur on the spot denigrating some aspect of your person that you do not find offensive (eg. Flumplenook - for a person who’s a bit clumsy), is it really a slur?

So if you call someone retarded, and they do not have the mental faculties to be offended, is it really a slur?

For slurs to have any meaning, any power, they need to be understood and internalized as offensive.

Only because they massively displaced a shitload of local business. Same with Amazon. If you have very little skills, where else are you going to work?

Rabbits aren’t salads, silly. They are second order salads.

I really enjoyed the game until The Event. I played a few more loops and was constantly irritated at The Event getting in the way. Like, I get it. I understand that is the point. It just ruined it for me. I don’t want to race a clock when I am exploring.

Well, it certainly ain’t Jables.

During my early adult years when I first moved out on my own and it was just me, I flipped my schedule to sleep 1700 until whenever I woke up. No alarms. Could sleep in every day because the result was “Oh no, still have many hours until work”. Would work 0700 until 1600. It was amazing. I was so awake and focused on my own stuff. Could practice piano, write poetry, work on open source code during those wee hours. Early morning work was also very productive. Afternoon work time was meh, but that was okay because of how the work was structured. Would bike into the office since it was only about 8km (5 mile) via residential streets. Do my grocery shopping at a 24hr market. Laundry room at my apartment complex was always open. It was such a magical time. Lonely, but would see friends late nights as their shifts ended or the evening was just peaking. Plus all my internet friends on IRC from all over.

What I’ve done before when I feel like I don’t have time for stuff I want to do is shift your hours a little bit to give yourself more time before work. If you take your meds first thing in the morning, you’ll get to enjoy all of that focus to yourself first, then work gets the dregs (as is proper)

Unexpected KSP is unexpected!

Why is work so important for you? I think you’ll find that a large number of people simply go through the motions because the stakes are low and their lives outside of work are more interesting. To them, it is an exchange of labor (that isn’t valued anyway) for (not enough) money. Why push yourself at work when it simply doesn’t matter? And what will drive you nuts later is that people from that “lazy” group will eventually end up promoted over you. The work is ultimately inconsequential, but the relationships built matter.

I don’t really have an answer for you other than to introspect a little bit on your work ethic.

This is a solvable problem. Just make a LoRA of the Alice character. For modifications to the character, you might also need to make more LoRAs, but again totally doable. Then at runtime, you are just shuffling LoRAs when you need to generate.

You’re correct that it will struggle to give you exactly what you want because you need to have some “machine sympathy.” If you think in smaller steps and get the machine to do those smaller, more do-able steps, you can eventually accomplish the overall goal. It is the difference in asking a model to write a story versus asking it to first generate characters, a scenario, plot and then using that as context to write just a small part of the story. The first story will be bland and incoherent after awhile. The second, through better context control, will weave you a pretty consistent story.

These models are not magic (even though it feels like it). That they follow instructions at all is amazing, but they simply will not get the nuance of the overall picture and be able to accomplish it un-aided. If you think of them as natural language processors capable of simple, mechanical tasks and drive them mechanistically, you’ll get much better results.

Maybe the problem is that I’m too close to the specific problem. AI tooling might be better for open-ended or free-association “why not try glue on pizza” type discussions, but when you already know “send exactly 4-7-Q-unicorn emoji in this field or the transaction is converted from USD to KPW” having to coax the machine to come to that conclusion 100% of the time is harder than just doing it yourself.

I, too, work in fintech. I agree with this analysis. That said, we currently have a large mishmash of regexes doing classification and they aren’t bulletproof. It would be useful to see about using something like a fine-tuned BERT model for doing classification for transactions that passed through the regex net without getting classified. And the PoC would be would be just context stuffing some examples for a few-shot prompt of an LLM and a constrained grammar (just the classification, plz). Because our finance generalists basically have to do this same process, and it would be nice to augment their productivity with a hint: “The computer thinks it might be this kinda transaction”