No. I have never and do not use AI in my art and never will

• When I’m stuck in debugging and can’t think of what can go wrong, LLM chats are quite useful. I can ask for possibilities and often I find something meaningful that didn’t come to mind. These kind of things are hard to do with search engines. (If I’m debugging something unfamiliar this becomes very counterproductive though, as I can’t filter hallucinations by looking).
• A smart text formatter
• Simple bash one liner, boilerplate code generation. I tried it but non trivial/bit longer code generation again gets effectively slower as I find myself fixing/working around AI mistake quite often.

I’m a radiologist and our group uses an LLM tool to assist with generating reports on imaging studies. Our reports have a body that includes all of the imaging findings (which we dictate) and then a conclusion/summary calling out what is most important (and serving as a tl;dr for other physicians). The LLM tool analyzes the body to generate that summary of important findings. It certainly is not perfect and frequently requires some editing. Overall it is faster than me creating the summary each time though.

I’ve used llama3 to help me rewrite my ancient CV and gotten good results so far. I took what it suggested and changed things myself to make a bit more sense. I also use it to summarize things occasionally but that’s about it.

58 comments? Meh I will put this in LLM and have it summarise it in bullet points.

I think ChatGPT and other Ai can be a fantastic tool if you use it responsibly. It’s a great help for learning and practicing things. I’ve very recently made my first server and it’s great at answering all my simple questions that I sometimes feel hesitant to bother people with, and little things like that. Sometimes I’ll ask it to give me a simple kind of structure or bullet point list of topics I need to make sure to hit in my writing, or weirdly enough it’s pretty good at helping me with substitutions in recipes, or other little things like that. And I personally think that all of this is fine.

But I’m entirely against using it to create any kind of final product. Having it do any kind of actual final work is just stupid and lazy. I truly don’t think Ai is capable of making anything that’s worth peoples time really, and in the amount of time you’ll spend meticulously explaining everything that you want or need for whatever it is you have it generating for you, you could’ve just made it yourself and done a far better job. That’s where I draw the line. I don’t think Ai has to be inherently evil or anything because it can be a great tool, but you can’t rely on it to actually do things for you. Maybe others will disagree me I know many people especially in Fediverse circles are very very strongly anti Ai in all facets but that’s just my thoughts.

Everyday, the company I work for have all their code in SAS, I use our LLM to translate it to python. I also write my python scripts and ask the llm to refract it and optimize it. Sometimes it save me 2 seconds so I just use my code that is usually simple, but other times it saves me half an hour.

I basically use it on rare rare occasion to help get me “unstuck” with creative tasks, I don’t really use what it produces in the end, I wind up dismantling it entirely and rewriting it “properly” but it has a use you know?

I use it all the time, and not just for myself or for work. Yesterday I fed my son’s study guide into ChatGPT and had it create a CSV file with flash cards for Anki. It’s great at any kind of transformation / summarizing or picking out specific information.

When school sends me overly verbose messages about everything that’s going on I can feed the message into ChatGPT and have it create an ical file that has events for the important stuff that happens in school in in the coming week.

I used it to write a greeting card for my dad on his birthday (“I’m giving him X, these are his interests, give me ten suggestions for greeting cards”).

I have it explain the reasons behind news stories (by searching for previous information and relating it to the news story). I ask tons of questions about anything I wonder about in the world such as chemical processes, the differences between oil frying and air frying, finding scientific papers about specific things, how to factory reset my Bose headphones… the list goes on.

I started using Debian full-time a year and half ago. It was a very frustrating experience initially and I leaned on LLMs heavily for advice. It was pretty hit or miss initially, and still occasionally gives wrong advice, but it has become much more helpful as the models have progressed. I have been able to restore a broken bup backup, learned the innards of systemd, troubleshoot scripts not launching correctly, optimized my Wayland config, correct fstab boot errors, configure my openWRT router, etc. Obviously I can just blindly copy/paste, but because I ask questions and try and tie things together, I learn along the way as well.

Currently taking a stats course and use the paid version of Claude to check/correct my workings when I do an exercise outside of the course. Also, it’s great for explaining concepts in relatable terms. For example I was having trouble understanding confidence intervals, but told Claude to explain it using Steph Curry’s 3pt shooting % as example.

There are going to be a lot of people left behind because they haven’t kept up w/ the rate of progress and still see LLMs as they were when they first launched.

Much as I dislike praising AI, I must admit I got some good results using an AI powered search engine for academic articles to find sources for a term paper I’m writing for a seminar class I’m taking for my masters degree.

I would say that I have used an LLM for productive tasks unrelated to work. I run a superhero RPG weekly, and have been using Egyptian & north African myths as the origin for my monsters of the week. The LLM has taken my research and the monster-creating phase of my prep from being multiple hours to sometimes under one hour - I do confirm everything the LLM tells me with either Wikipedia or a museum. I can also use an LLM to generate exemplary images of the various monsters for my players, as a visual aid.

That means I have more time to focus on the “game” elements - like specific combats, available maps, and the like. I appreciate the acceleration it provides by being a combined natural-language search engine and summary tool. Frankly, if Ask Jeeves (aka ask(dot)com) was still good at parsing questions and providing clear results, I would be just as happy using it.

I don’t use AI for productive work, for the same reasons I don’t stir my soup with a dishrag.

Pretty good for recipes, tho’.

Pretty useful for software engineering, particularly helpful in writing a test suite, you still need to actually check the output though ofc

Also made use of it for writing my end of year review to solve the blank page problem, I find it a lot easier to edit down than starting HR stuff like that entirely from scratch

I use it as a glorified Google search since Google search is absolute dogshit these days. But that’s about it. ChatGPT is one of the most over hyped bullshit I’ve ever seen in my life.