Hey guys, quick question.
How many of you use separate AI platforms when you are doing deep thinking and deep research?
I came across this bias about a year ago.
I was doing some research, did a little double-checking, and realized something. Using the same AI for both thinking and research makes the results biased.
I kind of attribute it to a student grading their own test.
The same brain that wrote the answer is the one deciding if the answer is right.
So now I am curious where everyone here lands on this.
Are you splitting it across two tools, or running it all in one?
And if you run it all in one, have you ever caught it telling you what you wanted to hear?