tracyspcy@lemmy.ml to

ChatGPT@lemmy.mlEnglish · 1 year ago

Over just a few months, ChatGPT went from correctly answering a simple math problem 98% of the time to just 2%, study finds

8

69

Over just a few months, ChatGPT went from correctly answering a simple math problem 98% of the time to just 2%, study finds

tracyspcy@lemmy.ml to

ChatGPT@lemmy.mlEnglish · 1 year ago

8

Over just a few months, ChatGPT went from accurately answering a simple math problem 98% of the time to just 2%, study finds

The chatbot gave wildly different answers to the same math problem, with one version of ChatGPT even refusing to show how it came to its conclusion.

Chat

BaroqueInMind@kbin.social
link
fedilink
arrow-up
3·
1 year ago
Self censored ChatGPT messing itself up reminds me when Pokémon hurt themselves in its confusion messages.

ChatGPT@lemmy.ml

chatgpt@lemmy.ml

You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: [email protected]

A community to discuss about ChatGPT and AI. Not affiliated with OpenAI.

Rules:

NSFW must be marked as such
No porn
Don’t be toxic
jailbreaks are allowed for now at least
No trashposts.
Keep the memes to a minimum

Visibility: Public

This community can be federated to other instances and be posted/commented in by their users.

4 users / day
1 user / week
4 users / month
4 users / 6 months
1 local subscriber
1.48K subscribers
42 Posts
88 Comments
Modlog