this post was submitted on 22 Feb 2024
110 points (93.7% liked)

Technology

59596 readers
3444 users here now

This is a most excellent place for technology news and articles.


Our Rules


  1. Follow the lemmy.world rules.
  2. Only tech related content.
  3. Be excellent to each another!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots


founded 1 year ago
MODERATORS
all 16 comments
sorted by: hot top controversial new old
[–] dojan@lemmy.world 39 points 9 months ago* (last edited 9 months ago) (1 children)

Ah, it's ambigaus again, is it?

Trying to overcome the bias in your dataset by introducing a bias of your own won't work. You're just setting yourself up to get a ridiculously (unintentionally) racist image of an "Ethinically Ambigaus" Homer Simpson.

Edit: fixed fucked sentence. Dunno what drugs I was on when I wrote that.

[–] andrew_bidlaw@sh.itjust.works 20 points 9 months ago (1 children)

It's kind of touching LLM companies are unlikely to outstand human creativity at it's worst. They just bought reddit datasets, and on reddit we saw many generations and kinds of open nazi empathisers going as far as creating 'clown world' communities where they used baby speak, were frens and felt terminally angry at 'nosy people'. And now Google still needs a lot of people to either exclude an excessive ammounts of keywords\combinations or sterilize their datasets. Good thing they can't just push the no-responsibility card and there is an outrage about copyrights and this shit. No one asked them to join that race, they get what they deserve.

[–] PerogiBoi@lemmy.ca 11 points 9 months ago (1 children)

Frenworl and the whole “honk honk 🤡” rhetoric was so embarrassing. It really highlighted the kind of intellect these jokers have.

[–] andrew_bidlaw@sh.itjust.works 5 points 9 months ago (2 children)

Yeah, right. I think it should be placed in some Internet Museum as one of the awkwardest points of it existence. They were ready to degrade themselves to that level just to keep being idiots. Landing there from links on other subs was a journey into some weird parallel reality.

[–] PerogiBoi@lemmy.ca 4 points 9 months ago

They’re a complete parody. It’s so easy to blend in with them and get them to do stupid things.

“Honk honk fren this thread is glowing. Must be from that sub! The wall is gonna be 10 feet higher now, pede! This train has no brakes. USA! Molon lube or something.”

[–] yamanii@lemmy.world 3 points 9 months ago (1 children)

Damn right person with a clown pfp.

[–] andrew_bidlaw@sh.itjust.works 9 points 9 months ago

I need to conserve an ideological purity of my clown craft.

[–] MysticKetchup@lemmy.world 18 points 9 months ago (1 children)

So is there an "AI is going just great" website yet like there is for Web3?

[–] SparrowRanjitScaur@lemmy.world -2 points 9 months ago (2 children)

AI is going great though, unlike web3. ChatGPT has already achieved mainstream adoption at a large number of major companies. And there were ads using images generated with midjourney at the super bowl.

[–] MysticKetchup@lemmy.world 12 points 9 months ago (1 children)
  1. You're literally commenting on an article about AI image generation having to be shut down because it's not going well
  2. Pets.com also advertised at the Super Bowl, it doesn't mean much other than spending a lot of money
[–] SparrowRanjitScaur@lemmy.world -1 points 9 months ago

I like how you didn't respond to the points I made about either ChatGPT or midjourney. ChatGPT has a number of large corporate customers now that use it as a tool in the creation of professional products. And midjourney is still going strong and is widely used, which is why I mentioned the Superbowl.

Yes, gemini images are on pause until they fix the vulnerabilities to abuse. But I use Gemini as the engine for Google assistant on my phone now, so the service is obviously still gaining traction.

[–] HarkMahlberg@kbin.social 10 points 9 months ago (2 children)
[–] dojan@lemmy.world 7 points 9 months ago

Air Canada had to shut down their AI chatbot because it was making up refund policies and costing them money.

HA. I fucking love that.

[–] Klicnik@sh.itjust.works 3 points 9 months ago (1 children)

Duolingo has dropped hugely in quality (for Korean anyway) just in the last couple months. The AI-generated voices mispronounce so much stuff. I spend more time reporting unclear audio or outright mispronunciations and grammar issues than I do learning. They seem to want their users to be their QA team, but then don't even fix the reported issues. https://www.msn.com/en-us/news/other/duolingo-lays-off-workers-as-it-leans-on-ai-tools-to-carry-out-more-tasks/ar-AA1mKgja

[–] HarkMahlberg@kbin.social 3 points 9 months ago

Poor Duolingo. Once upon a time I used it to learn Japanese, but by the time I could start reading kanji and noticed that duolingo was still constructing sentences entirely out of hiragana, I knew I had outgrown it and moved on to Anki.

Using AI to learn a new language has to be incredibly frustrating - you can either tell where's messing up, or you can't tell at all and then you learn incorrect information..