Did Bing and Gemini Got Dumbed Down Recently?

Posted by xchip on Monday April 14, 2025 @08:07PM
from the self-post dept.

I tried to make it compare features from different products and they were making lots of mistakes.

Anyone is getting the same impression that AI models are worse now than before?

( 5 comments | on HN )

Re:Did Bing and Gemini Got Dumbed Down Recently? by bigyabai on Monday April 14, 2025 @08:08PM (#43687663)

They've never worked for me in the first place

[ Parent | on HN ]

Re:Did Bing and Gemini Got Dumbed Down Recently? by menshiki on Monday April 14, 2025 @09:53PM (#43688258)

I've experienced the opposite. Recently, I've been getting amazing results from Gemini running 2.5 Pro.

[ Parent | on HN ]

Re:Did Bing and Gemini Got Dumbed Down Recently? by dantodor on Monday April 14, 2025 @10:03PM (#43688321)

Gemini for sure, in my experience. I asked 2.5 (presumably the most advanced in the known Universe) to write some code. First it came back with suggestions on HOW to write that code. Asked again. He wrote some, but from 4 functions implemented, 2 were just comments explaining again how to implement that function. Notable mention, one of them was supposed to call the Gemini API. Everything Python, nothing fancy. Took me like 6 rounds of back and forth until I got some very basic structure... So, no vibe coding with Gemini for me.

[ Parent | on HN ]

Re:Did Bing and Gemini Got Dumbed Down Recently? by dalmo3 on Wednesday April 16, 2025 @01:42AM (#43701878)

Ha, it really goes wild on comments. Just seen it write a 5 LOC function, with `return` and all, then add 15 lines of comments second guessing itself, then writing the same 5 LOC differently... In the same function body.

Fraternal twins for sure.

[ Parent | on HN ]

Re:Did Bing and Gemini Got Dumbed Down Recently? by runjake on Tuesday April 15, 2025 @12:54PM (#43695482)

For Gemini, I’m seeing the opposite. It’s been getting much better.

[ Parent | on HN ]