I have that almost every game, or I use my time to get an even position and end up losing on time. but I also have good bullet skills and so I'm able to convert a lot of the time. but Im trying to get better at spending less time on the less decisive moves and more time on the more important moves like where there's a tactic or something like that.
good luck getting to 3000! 
I think you're over complicating it, simplify things more and just concentrate on winning the whole game rather than splitting it in two, if you're winning positionally and up material entering into time scrambles just makes it a lottery as the opponent will more than likely spam pre moves if in a losing position
I'm not sure if you can trust what chatgpt says.
I also like blitz and is one of my strong spots but I usually try to play quicker than my opponent and make moves that confuse him (I try to avoid trades) and at the end I make some unexpected blunders that check the opposing King and win on time
ChatGPT sucks and you're right, I also use this strategy (although in my patzer level lmao)
hello friends! <3 I was discussing with chatgpt(AI) about playing styles in blitz because I want to push closer to 3000 this year, and I came across an interesting paradox that I would like to share with you. in blitz there are 2 main factors: the position and the clock. Both are nearly equally important because the position determines who is going to win the game, but also none of that matters if you run out of time on the clock. And normally what I would do in my own games is use nearly all of my time to get a completely winning position, essentially relying on my bullet skills to convert the game afterward quickly, gaining the ability to get extremely low on time like many players cannot allow themselves to do without losing on time. Finishing the game before my clock runs out. This normally works for me. I have been practicing a lot with my mouse speed lately and haven't had much of a problem with it. But chatgpt(AI) suggested that I was causing my win ratio to significantly decrease by doing this, because I first have to win the game on the board, but then I have to make it more complicated by winning yet again in a scramble. I’d be interested to hear if anyone else ends up “winning the game twice” like this, or if this is actually a major hole in my approach to blitz chess altogether. And how can I go from winning two games in one to one game per game?