Gemini - Confidently Wrong

I saw an interesting comparison between Gemini Ultra, Claude 2.1 (no idea what this is), and ChatGPT 4 on Reddit recently. It pointed out the fact that Gemini has a tendency to be confidently wrong, and I decided to test this out for myself.

The prompt in question is - Is a kilo of feathers heavier than a pound of steel? Spoiler alert, only ChatGPT 4 seemed to provide a sensible answer.

Here is ChatGPT's response:

1 kilogram is indeed approximately 2.2 pounds so I'd say ChatGPT's response is spot on (unless I'm missing something).

Here is Gemini's (not Ultra) response:

When I asked Gemini to double-check, it course-corrected with a long-winded explanation.

I think Gemini still has a strong potential to be a ChatGPT alternative in the future, but accuracy/correctness remains a weaknesses. 

This appeared to be the case when I asked Gemini about FX brokers that accepted Canadian residents since I know a handful do. Gemini listed some brokers such as TMGM, which I've never heard of and, to my knowledge, does not comply with Canadian regulatory requirements.

As always, responses should always be taken with a grain of salt regardless of the chatbot solution used.