45
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 17 Apr 2024
45 points (100.0% liked)
technology
23313 readers
39 users here now
On the road to fully automated luxury gay space communism.
Spreading Linux propaganda since 2020
- Ways to run Microsoft/Adobe and more on Linux
- The Ultimate FOSS Guide For Android
- Great libre software on Windows
- Hey you, the lib still using Chrome. Read this post!
Rules:
- 1. Obviously abide by the sitewide code of conduct. Bigotry will be met with an immediate ban
- 2. This community is about technology. Offtopic is permitted as long as it is kept in the comment sections
- 3. Although this is not /c/libre, FOSS related posting is tolerated, and even welcome in the case of effort posts
- 4. We believe technology should be liberating. As such, avoid promoting proprietary and/or bourgeois technology
- 5. Explanatory posts to correct the potential mistakes a comrade made in a post of their own are allowed, as long as they remain respectful
- 6. No crypto (Bitcoin, NFT, etc.) speculation, unless it is purely informative and not too cringe
- 7. Absolutely no tech bro shit. If you have a good opinion of Silicon Valley billionaires please manifest yourself so we can ban you.
founded 4 years ago
MODERATORS
And this is a change due to using an LLM? Assuming so, I'm absolutely astonished. Can I ask what work the quicker workers do, and how an LLM has helped them to be quicker?
if your job is generating bullshit then the bullshit generators are a godsend
Emails
a lot of my job is 'emails' but I can't just use a LLM for responses - it'd be so obviously bad and hallucinate solutions to problems that don't exist. god I wish.
If you have enough emails from your job you could fine-tune one of the models on them and it would probably do pretty okay. It'd be really cheap cus it's not that much data, assuming you don't have the hardware to do it yourself.
It would sound like me, sure. It would generate emails saying that a course of action will or won't succeed. The main difference would be that my emails are informed by investigating things often not captured in the email chain to which I'm replying. robo-ped_xing would be just as confident as me, much cheaper, faster and utterly wrong very often. If that's an OK trade-off, they could have replaced me with a magic 8-ball years ago.
then your job isn't just emails, it's the independent research and knowledge/experience.
But yeah, your job has to be basically pure bullshit for this to work.
But do you need to come up with the right answer to be paid?
It's kinda like kicking the extra point in American football. I don't have to get it right 100% of the time but I'm sure I'd be looking for work if my average were significantly lower.
Yeah, I could easily train a model on the multiple evernote notebooks I've accumulated over the years but to be honest it'd just basically replace me copying & pasting from Evernote and not actually save me any significant time on the brunt of my job - which is usually solving new and unique problems via email that the software engineers cause with every release.
It's the CEO, probably.
Actually. Fucking emails
Thanks for your response. Alas, I'm still completely oblivious as to how ChatGPT helps you work quicker.
Most emails are formulaic bullshit. The exact type of writing llms were designed for.