this post was submitted on 07 Mar 2024
486 points (97.5% liked)
Technology
59119 readers
2230 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Good. I hope that once companies stop putting AI in everything because it's no longer profitable the people who can actually develop some good tech with this can finally do so. I have already seen this play out with crypto and then NFTs, this is no different.
Once the hype around being able to make worse art with plagiarised materials and talking to a chatbot that makes shit up died down companies looking to cash out with the trend will move on.
At one point I agreed but not anymore. AI is getting better by the day and already is useful for tons of industries. It's only going to grow and become smarter. Estimations already expect most energy producted around the world will go to AI in our lifetime.
The current LLM version of AI is useful in some niche industries where finding specific patterns is useful but how it's currently popularised is the exact opposite of where it's useful. A very obvious example is how it's accelerating search engines becoming useless, it's already hard to find accurate info due the overwhelming amount of AI generated articles with false info.
Also how is it a good thing that most energy will go to AI?
LLMs should absolutely not be used for things like customer support, that's the easiest way to give customers wrong info and aggregate them. For reviewing documents LLMs have been abysmally bad.
For gammer it can be useful but what it actually is best for is for example biochemistry for things like molecular analysis and creating protein structures.
I work in an office job that has tried to incorporate AI but so far it has been a miserable failure except for analysing trends in statistics.
A LLM is terrible for molecular analysis, AI can be used but not LLM.
AI doesn't exist currently, that's what LLMs are currently called. Also they have been successfully used for this and show great promise so far, unlike the hallucinating chatbot.
AGI Artificial General Intelligence doesn't exist that is what people think of in sci-fi like Data or Hal. LLM or Large Language Models like CHAT GPT are the hallucinating chat bots, they are just more convincing than the previous generations. There are lots of other AI models that have been used for years to solve large data problems.
Pretty much anything Google is giving me says they are using deep learning LLMs in biology.
I agree about customer support, but in the end it's going to come down to number of cases like this, how much they cost, versus the cost of a room of paid employees answering them.
It's going to take actual laws forbidding it to make them stop.
Oh, yea, of course companies will take advantage of this to just replace a ton of people with a zero cost alternative. I'm just saying that's not where it should be used as it's terrible at those tasks.