this post was submitted on 02 May 2025
572 points (95.8% liked)

Technology

69771 readers
3966 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 news or articles.
  3. Be excellent to each other!
  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, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
  9. Check for duplicates before posting, duplicates may be removed
  10. Accounts 7 days and younger will have their posts automatically removed.

Approved Bots

founded 2 years ago
MODERATORS
L3s@lemmy.world
enu@lemmy.world
technopagan@lemmy.world
L4s@lemmy.world
L3s@hackingne.ws
L4s@hackingne.ws
572
AI models routinely lie when honesty conflicts with their goals (www.theregister.com)
submitted 4 days ago by cm0002@lemmy.world to c/technology@lemmy.world
118 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] FreedomAdvocate@lemmy.net.au 3 points 4 days ago (1 children)

It completed the task, it was just wrong.

[–] thedruid@lemmy.world -5 points 4 days ago (2 children)

No. It gave the wrong answer therefore it didn't complete the task. It gave the wrong answer. Task incomplete

That's literally how a task works.

[–] FreedomAdvocate@lemmy.net.au 1 points 19 hours ago (1 children)

The task was to tell you the difference. It told you the difference, it was just wrong about the differences. Giving an incorrect answer to a task is still doing the task.

[–] thedruid@lemmy.world 0 points 14 hours ago (1 children)

I swear. The mental gymnastics

If I tell you to take the garbage out, then you throw it through the window , that isn't molesting the task. Anymore than this does

1+1 is 2. Don't freaking force crap to fit into that "feature, not a bug" bullshit. It's the issue we have with products in general.

Excuses instead of results. Welcome to the world of a. I support

[–] FreedomAdvocate@lemmy.net.au 1 points 5 hours ago (1 children)

So if you answer a question wrong on a test you didn’t complete the test?

[–] thedruid@lemmy.world 0 points 1 hour ago

No. You didn't. you completed the percentage of the test that you answered correctly. Well supposedly.

Bottom line. A. I. isn't ready at all. Especially if one has to argue to make it's failures seem like a success

[–] Chozo@fedia.io 5 points 4 days ago (1 children)

Bruh. The task was to mislead. That's what it did.

[–] thedruid@lemmy.world 0 points 4 days ago (1 children)

That's not the task I mentioned.

I asked what the difference between two modes is. It was wrong. It was confidently wrong.

I

[–] Chozo@fedia.io 3 points 4 days ago (1 children)

Are you talking about the hypothetical scenario you made up about the subject you don't even understand? No wonder we're going in circles.

I'm talking about the article.

[–] thedruid@lemmy.world -1 points 3 days ago (1 children)

Did you go to trump university? You responded to MY comments which were clear.

Stop moving goal posts ,making assumptions, and resorting to name-calling g it's a sign of stupidity, that I am not ready to truly apply to you, yet

Now. Good evening.

[–] Chozo@fedia.io 2 points 3 days ago

Nobody called you anything, dude. Are we having different conversations or something?