53

Benchmarking AIs with a simple riddle(benchmarks.llmonitor.com)

posted
by
Avatar
marco@beehaw.org
in
Avatar
technology@beehaw.org
6 comments
hidereport

Sally (a girl) has 3 brothers. Each brother has 2 sisters. How many sisters does Sally have?

The correct answer is 1 - slightly more compared to how many AIs got it right.

Sort:
HotTopControversialNewOld
Avatar
raccoona_nongrata@beehaw.org
19 points
Deleted by creator
permalink
report
reply
Avatar
ceuk@feddit.uk
6 points
permalink
report
parent
reply
Avatar
marco@beehaw.orgOP
3 points

I wonder if someone told CGPT ;)

permalink
report
parent
reply
Avatar
Jaysyn@kbin.social
7 points

ChatGTP got it first try for me

“Sally is one of the sisters mentioned in the statement, so she has one sister. The statement doesn’t specify any other sisters, so Sally has one sister.”

permalink
report
reply
Avatar
JonC@programming.dev
3 points
*

Same, using Chat GPT 4. It explained the steps without prompting, which is different from the single line answer shown in the post too. I got this…

Let’s break this down step by step:

  1. Sally has 3 brothers.
  2. Each of those brothers has 2 sisters.

Sally is one of those sisters for each of her 3 brothers. Therefore, the second sister that each brother has would be the same other sister.

This means that Sally has only 1 other sister, making a total of 2 sisters in the family (including Sally herself).

So, Sally has 1 sister.

permalink
report
parent
reply
Avatar
drifty@lemmy.ml
2 points

do you mean GPT*?

permalink
report
parent
reply
Avatar
user@lemmy.one
1 point

good 1. ai.ls got it right. the element/riot group chatgpt 3.5 got it wrong lol. http://0x0.st/Hf36.png

permalink
report
parent
reply

Technology

!technology@beehaw.org

Create post

A nice place to discuss rumors, happenings, innovations, and challenges in the technology sphere. We also welcome discussions on the intersections of technology and society. If it’s technological news or discussion of technology, it probably belongs here.

Remember the overriding ethos on Beehaw: Be(e) Nice. Each user you encounter here is a person, and should be treated with kindness (even if they’re wrong, or use a Linux distro you don’t like). Personal attacks will not be tolerated.

Subcommunities on Beehaw:

This community’s icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.

Community stats

  • 2.7K

    Monthly active users

  • 3.5K

    Posts

  • 82K

    Comments

Community moderators